Predicting Patient Advocacy Engagement: A Multiple Regression Analysis Using Data From Health Professionals in Acute-Care Hospitals

Although literature documents the need for hospital social workers, nurses, and medical residents to engage in patient advocacy, little information exists about what predicts the extent they do so. This study aims to identify predictors of health professionals’ patient advocacy engagement with respect to a broad range of patients’ problems. A cross-sectional research design was employed with a sample of 94 social workers, 97 nurses, and 104 medical residents recruited from eight hospitals in Los Angeles. Bivariate correlations explored whether seven scales (Patient Advocacy Eagerness, Ethical Commitment, Skills, Tangible Support, Organizational Receptivity, Belief Other Professionals Engage, and Belief the Hospital Empowers Patients) were associated with patient advocacy engagement, measured by the validated Patient Advocacy Engagement Scale. Regression analysis examined whether these scales, when controlling for sociodemographic and setting variables, predicted patient advocacy engagement. While all seven predictor scales were significantly associated with patient advocacy engagement in correlational analyses, only Eagerness, Skills, and Belief the Hospital Empowers Patients predicted patient advocacy engagement in regression analyses. Additionally, younger professionals engaged in higher levels of patient advocacy than older professionals, and social workers engaged in greater patient advocacy than nurses. Limitations and the utility of these findings for acute-care hospitals are discussed.


INTRODUCTION
This study addresses a large gap in existing research literature: the paucity of information about what predicts health professionals' engagement in patient advocacy. Without this knowledge, it is not possible to remedy the barriers that might exist to their provision of it or to develop methods of increasing health professionals' engagement in patient advocacy, such as through organizational changes or advocacy training.
To address this gap, the current study sought to specify scales that predicted health professionals' levels of patient advocacy engagement in acute-care hospitals. The scales were conceptualized and developed by the research team in collaboration with project stakeholders (discussed below), and were chosen for inclusion in the current study based on a thorough review of the literature that indicated their relevance to the topic. Specifically, Azjen's (1985) Theory of Planned Behavior, which asserts that people intend to and do engage in an activity when they feel driven or eager to do so, prompted the development of a scale to measure health professionals' eagerness to engage in patient advocacy ("Eagerness"). Since many theorists have linked patient advocacy to the ethical beliefs of health care professionals, we developed a scale to measure their ethical commitment to patient advocacy ("Ethical Commitment") (Bu & Wu, 2008;Dodd, Jansson, Brown-Saltzman, Shirk, & Wunch, 2004;Earnest, Wong, & Federico, 2010;Grady et al., 2008;Hanks, 2008;Sundin-Huard & Fahy, 1999). Literature that suggests health professionals are more likely to engage in patient advocacy if they possess specific skills to do so indicated the need for a scale to measure their patient advocacy skills ("Skills") (Altun & Ersoy, 2003;Dodd et al., 2004;Hanks, 2008;Itzhaky, Gerber, & Dekel, 2004;Stafford, Sedlak, Fok, & Wong, 2010;Ware, Bruckenthal, Davis, & O'Connor-Von, 2011). Research documents that employees engage in activities perceived to have few obstacles and that are associated with resources and opportunities to engage in them (Armitage & Conner, 2001;Sellin, 1995); thus we developed a scale to measure the extent health professionals receive tangible job supports to engage in patient advocacy ("Tangible Support"). Work environment and working conditions that impact patient advocacy have been widely discussed, and thus led to the development of a scale to measure the hospital's organizational receptivity to patient advocacy ("Organizational Receptivity") (Brown & Leigh, 1996;Chafey, Rhea, Shannon, & Spencer, 1998;Josse-Eklund, Petzäll, Sandin-Bojö, & Wilde-Larsson, 2013;Sundin-Huard & Fahy, 1999). Theory indicating that individuals are more likely to engage in a behavior when they are subject to social pressures to engage in it (Azjen, 1985) signaled the need for a scale to measure the extent other professionals are perceived to engage in patient advocacy ("Belief Colleagues Engage"). Finally, as Makary (2012) argues that health professionals are more likely to engage in patient advocacy if their hospital's culture emphasizes patient empowerment, we developed a scale that measures the extent health professionals believe their hospital empowers patients to advocate for themselves ("Belief the Hospital Empowers Patients"). The current study assesses the extent to which these predictor scales are associated with health professionals' engagement in patient advocacy through bivariate and multiple regression analyses.
In the United States, social workers, nurses, and medical residents are required by their Codes of Ethics to engage in patient advocacy. The Standards for Social Work Practice in Health Care Settings state that social workers "have a responsibility to advocate for the needs and interests of clients and client systems in health care" (National Association of Social Workers, 2005, p. 24). The Code of Ethics of the American Nurses Association (2014) states, "The nurse promotes, advocates for, and strives to protect the health, safety, and rights of the patient" (p. 1). The preamble of the American Medical Association's (2014) Code of Ethics states that physicians should "recognize a responsibility to seek changes in those requirements which are contrary to the best interests of the patient" and "participate in activities contributing to the improvement of the community and the betterment of public health" (para. 4, 8). The Joint Commission requires hospitals to develop written policies on an array of stated patient rights and to put respect for the patient's rights into action in myriad, concrete ways (Joint Commission, 2009). If health professionals fail to detect and advocate to resolve patients' problems, these problems may fester and lead to negative outcomes, including patient dissatisfaction, poorer health outcomes, litigation, and increased cost of care.
Considerable literature emphasizes barriers to engagement in patient advocacy. Hard-pressed health professionals often lack time and skills to provide effective patient advocacy (Davis et al., 2003;Donaldson, 2007;Faust, 2008;Herbert & Levin, 1996;Itzhaky et al., 2004;Foley, Minick & Kee, 2002;Weiss-Gal & Gal, 2009). Fear of disrupting working relationships with physicians, becoming unpopular, being labeled a troublemaker, and exposing oneself to attack are barriers to engagement in patient advocacy by some nurses (Sellin, 1995). Lack of autonomy, fatigue, power hierarchies, and physician demeanor have been identified as additional barriers to engaging in patient advocacy (Chafey et al., 1998) Less is known about what factors facilitate or contribute to engagement in patient advocacy by frontline health professionals. Kubsch, Sternard, Hovarter, and Matzke (2004) explored whether nurses' moral development, assertiveness, and job security influenced their engagement in different forms of advocacy, including moral-ethical, legal, political, spiritual, and substitutive advocacy. Moral stage development was found to have a significant effect on substitutive advocacy, but assertiveness and job security were not significant factors influencing any category of advocacy (Kubsch et al., 2004). Dodd et al. (2004) explored factors that predicted nurses' engagement in advocacy pertaining to patients' ethical issues. The extent they felt included in ethics deliberations and their level of ethics education were significant predictors of their advocacy engagement (Dodd et al., 2004)

Focus of the Current Study
The current study draws on Jansson's (2011) definition of patient advocacy: "An intervention to help patients obtain services, rights, and benefits that would (likely) not otherwise be received by them and that would advance their wellbeing" (p. 4). We expanded the definition to include the following caveat: "Patient advocacy can be provided directly to patients or through referrals provided that health professionals ascertain if patients actually received assistance." This definition usefully precludes relatively minor instances of advocacy by only including advocacy interventions that health professionals' believe are essential to patients' well-being, only when they believe no one else will step forward to help the patient, and only when they follow-up on their referrals to be certain that patients actually receive assistance. Further, patient advocacy was conceptualized to apply to seven categories of patient problems identified by Jansson (2011): ethical rights, quality care, culturally competent care, preventive care, affordable care, mental health care, and care linked to patients' households and communities. To measure health professionals' level of engagement in patient advocacy, we utilized the validated Patient Advocacy Engagement Scale (Patient-AES; Jansson et al., 2015) in bivariate and multiple regression analyses, as specified below.

Study Aims
The research questions that guided this study were as follows: (1) To what extent are seven predictor scales (Eagerness, Ethical Commitment, Skills, Tangible Support, Organizational Receptivity, Belief Colleagues Engage, and Belief the Hospital Empowers Patients) associated with health professionals' levels of patient advocacy engagement and (2) To what extent do these seven predictor scales, when controlling for specific sociodemographic and setting variables, predict health professionals' levels of patient advocacy engagement?

Research Design
We used a cross-sectional research design with data collected from a sample of social workers, nurses, and medical residents in eight acute-care hospitals in Los Angeles to answer the research questions.

PROCEDURES
Survey Design. With the support of an expert stakeholder panel, the research team met in the summer of 2012 to consult the existing literature on advocacy in health care settings and to discuss the development of scales to measure health professionals' engagement in patient advocacy and its possible predictors. All scales, as well as many sociodemographic variables, were compiled into a Qualtrics survey containing more than 400 total items. For the current study, the Patient-AES was conceived as the dependent variable and the predictor scales, sociodemographic, and setting variables as independent variables.
The project's stakeholder panel included a social worker who supervised a hospital case management program for 20 years, a breast cancer survivor who successfully lobbied for state legislation to enhance the care of breast cancer patients with dense breast tissue, a physician who pioneered advocacy training for individuals with substance abuse withdrawal symptoms, a nurse who headed a university-based center on bioethics with expertise in patient advocacy for individuals at end-of-life, a social worker who pioneered advocacy for discharged patients at a major public hospital for 30 years, the head nurse of a major hospital who founded an annual award for nurses who excelled in patient advocacy, an Associate Dean for Research at a School of Nursing who had conducted extensive research related to advocacy for persons with HIV/AIDS, an Associate Professor of Social Work with research expertise on advocacy with respect to ethical issues, and a Clinical Associate Professor of Social Work with expertise in advocacy for senior citizens.
Hospital Selection. We obtained data from health professionals in eight acute-care hospitals of different types so that idiosyncratic characteristics of specific kinds of hospitals would not bias measurement of health professionals' responses to questions in the online survey. We also selected hospitals in which stakeholders and members of the research team had contacts to allow expeditious obtainment of Institutional Review Board (IRB) approvals. The eight participating hospitals included a community-based nonprofit hospital, a university-affiliated nonprofit general hospital, a public children's hospital, a public general hospital, a veterans' hospital, a university-affiliated cancer hospital, and two church-affiliated hospitals. We had initially planned to include a for-profit hospital but were unable to obtain requisite permission despite repeated efforts.
Participant Eligibility. A power analysis indicated that approximately 300 participants were needed to ensure sufficient effect size for statistical analyses with roughly 100 respondents from each of the three professional groups in order to enable comparisons. Health professionals who had served at least six months in their hospitals were sought to enhance the likelihood they would be familiar with its personnel and policies. The research team 564 B. S. Jansson et al. selected nurses, social workers, and medical residents because they typically are positioned to serve large numbers of patients in their hospitals rather than specific caseloads often served by attending and consulting physicians. Nurses, social workers, and medical residents can act as "case finders" as they make rounds within their respective units. Medical residents are taught to identify patients with unresolved problems as part of their medical training.
No restrictions were placed on medical residents' area of specialty such as their unit or area of specialization.
Inclusion criteria for the sample included the following: (1) participant must work full time, part time, or per diem and have worked in this hospital setting for at least six months and (2) participant must be a social worker, nurse, or medical resident in the hospital. Nurses were minimally required to have an RN degree and social workers were required to have an MSW. Temporary nurses and social workers, as well as student workers were excluded.
Participant Recruitment. Staff rosters of all social workers, nurses, and medical residents were obtained at each participating hospital. All social workers were contacted in each participating hospital because their rosters were small. A random number generator was used to generate a pool of nurses and medical residents from the rosters of the eight hospitals. These individuals were contacted via e-mail and provided with information about the study and a link to the online survey. Participation was voluntary. Response rates varied by site and profession. Of the 732 total professionals invited to participate, 40% consented to participate and completed the online survey. Fifteen individuals were ineligible for the study and an additional 29 started but did not complete the survey. For a full report of the number of social workers, nurses, and medical residents invited to participate in each of the eight hospitals, the number who completed the survey, and the response rate for each profession, see Jansson et al., 2015. Data Collection. The online Qualtrics survey was launched in September 2013 with data collection taking place over the following five months. Participants were given one month to complete the survey once they started it, which they could leave and resume at any point during the month. Participants received $100 after completing the survey, which contained more than 400 total items and which took 35 minutes on average to complete. The study was approved by the IRBs of all participating hospitals, as well as by the University of Southern California.

MEASURES
This section describes the independent and dependent variables that were utilized in the bivariate and multiple regression analyses. Items within the Patient Advocacy Engagement Scale (Patient-AES), as well as all  Have you been excluded from discussions about unresolved patient problems? (reverse coded) Are you invited to participate in case conferences about patients with specific unresolved problems? Is there an atmosphere that invites you to question the resolution of unresolved problems with specific patients? Do you experience hostile behaviors from other professionals? (reverse coded) Is sufficient discussion devoted to patients' unresolved problems in your setting by health care professionals, supervisors, and administrators? Do patients' attending or consulting physicians ask you to gather information from patients relevant to specific unresolved problems? Do patients' attending or consulting physicians ask you to gather information from family members, friends, or significant others that is connected to patients' unresolved problems? Do patients' attending or consulting physicians confidentially discuss with you patients' unresolved problems when several courses of action exist in a specific case? Do patients' attending or consulting physicians encourage you to inform them when you see patients' unresolved problems? Are you encouraged by attending or consulting physicians to become involved in a range of patients' unresolved problems? Are members of your profession encouraged to make referrals to ethics committees related to one of the unresolved problems?
Predicting Patient Advocacy Engagement predictor scales, were operationalized with response sets ranging from 1 (never) to 5 (almost always) or from 1 (not at all) to 5 (a great deal), depending on the nature of the question. Items contained within each of the scales are presented in Table 1. Dependent Variable. The Patient-AES is a 26-item instrument designed by Jansson et al. (2015) to measure the frequency with which health care professionals engage in patient advocacy related to specific patient problems in seven categories: (1) patients' rights, (2) quality care, (3) culturally Are views of members of your profession solicited by ethics committees with regard to unresolved problems? Are specific kinds of patients' unresolved problems addressed by multidisciplinary teams (e.g., physicians, social workers, and nurses)? Are you an integral member of discussions about patients' unresolved problems? When nurses, social workers, and medical residents participate in discussion about patients' unresolved problems, do they participate as co-equals? Patient Advocacy Tangible Support Scale ("Tangible Support") "To what extent": Do you believe administrators are aware of unresolved patient problems? Does your supervisor encourage you to engage in patient advocacy? Does your supervisor support you when your patient advocacy results in negative repercussions? Do you fear supervisors will not come to your defense when you are criticized for patient advocacy? (reverse coded) Is patient advocacy part of your job description? Have cuts in your profession in the past three years curtailed patient advocacy in your hospital? Is patient advocacy discussed during orientation of new staff? Is patient advocacy emphasized in your hospital's mission? Does your hospital emphasize patient advocacy compared to other hospitals? Is patient advocacy discussed in informational or educational materials provided to patients? Belief Colleagues Engage in Advocacy Scale ("Belief Colleagues Engage") How confident are you that patients with serious unresolved problems will receive patient advocacy in this hospital? What extent do you believe that the social workers engage in patient advocacy in your hospital? What extent do you believe that the nurses engage in patient advocacy in your hospital? What extent do you believe that the medical residents engage in patient advocacy in your hospital? What extent do you believe that the physicians (other than medical residents) engage in patient advocacy in your hospital?  (7) community-based care. Jansson et al. (2015) report data that supports both the validity and reliability of the Patient-AES as a multidimensional scale for measuring the frequency of advocacy engagement by social workers, nurses, and medical residents in acute-care settings. Cronbach's α for the Patient-AES was 0.95. Independent Variables. The independent variables included the following: Age. Participants were asked to indicate their age in years. This data was used to create a continuous variable.
Race. Participants were asked to self-identify as Caucasian/White, Hispanic/Latino, Asian/Pacific Islander, African American, Middle Eastern/Arab, American Indian/Alaskan Native, or "Other." These data were used to create a nominal variable with just three categories-White, Asian, and Other-since so few respondents identified as other than White or Asian.
Gender. Participants were asked to self-identify as male or female. This data was used to create a dichotomous variable.
Hospital Site. Participants were asked to indicate in which of the eight participating hospitals they were employed. This data was used to create a nominal variable with eight categories. This variable was included in the regression analysis to test whether the specific type of hospital, such as veterans', children's, or cancer hospitals, influences the extent of patient advocacy engagement. Per our IRB protocol, we are unable to provide the names of the specific hospitals, thus they are labeled as Site 1, Site 2, and so forth.
Profession. Participants were asked to indicate their profession-social worker, nurse, or medical resident. These data were used to create a nominal variable with three categories.
Patient Advocacy Eagerness. We developed a scale that measures respondents' eagerness to engage in higher levels of patient advocacy in the future by asking the extent over the past two months they wished they had engaged in more advocacy with respect to seven categories of patient problems ("Eagerness"). Cronbach's alpha for the scale was 0.86.
Patient Advocacy Ethical Commitment. We developed a 9-item scale to measure the extent respondents are ethically committed to patient advocacy ("Ethical Commitment"). Cronbach's alpha for the scale was 0.87.
Patient Advocacy Skills. We developed a 16-item scale designed to measure the extent respondents feel they possess skills necessary to engage in patient advocacy ("Skills"). Cronbach's alpha for the scale was 0.95.
Patient Advocacy Tangible Support. We developed a 10-item scale to measure the extent respondents report they receive tangible support for patient advocacy ("Tangible Support"). Cronbach's alpha for the scale was 0.82.
Organizational receptivity. We developed a 15-item scale to measure the extent a hospital's organizational climate supports health professionals' engagement in patient advocacy ("Organizational Receptivity"). Cronbach's alpha for the scale was 0.91.
Belief Colleagues Engage in Patient Advocacy. We developed a 5-item scale to measure the extent respondents believe that their professional colleagues (i.e., social workers, nurses, medical residents, and other physicians) engage in patient advocacy. It included an item that measures respondents' level of confidence that patients with serious unresolved problems will receive advocacy in their hospital in general ("Belief Colleagues Engage"). Cronbach's alpha for the scale was 0.74.
Belief the Hospital Empowers Patients. We developed a 7-item scale to measure the extent respondents believe that a culture of empowerment exists in their hospitals. Items asked respondents to rank the extent their hospital's professionals empower patients in various ways, including referring them to the Internet, helping them obtain second opinions, and directing them to patients with similar health problems ("Belief the Hospital Empowers Patients"). Cronbach's alpha for the scale was 0.83.

Statistical Analysis
Descriptive statistics, such as means on the scale variables and frequencies on sociodemographic and setting variables were obtained for the overall sample. Cronbach's alpha was calculated for each of the predictor scales used in the study. A bivariate analysis was then conducted examining the Pearson intercorrelations among the Patient-AES and the predictor scales. Bivariate analysis also involved the intercorrelations of the predictor scales to assess their multicollinearity. Next, a multiple regression analysis was conducted to assess the extent each of the predictor scales, along with the sociodemographic and settings variables, was independently related to the Patient-AES adjusting for all other independent variables.
The bivariate analysis indicated a degree of multicollinearity and a variance inflation factor was introduced. Highly intercorrelated predictor scales were retained in the analysis because the research question focused on the extent that each of the theoretically meaningful predictor scales contributed to the variation in the levels of patient advocacy engagement, as measured by the Patient-AES. Unstandardized betas for each independent variable were analyzed in order to determine precisely the level of change in the dependent variable accounted for by a change in the independent variable. The overall R-square of the regression model was calculated in order to assess the percentage of the variance in the Patient-AES that was explained by the independent variables. All analyses were performed in SAS (v 9.4).

Multiple Regression
The multiple regression analysis determined that the model consisting of predictor scales, sociodemographic and setting variables explained 40% of the variance in the Patient-AES. Younger age (unstandardized β = -0.31, p < .01) significantly predicted higher engagement in patient advocacy, as measured with the Patient-AES. Gender and Race were not significantly related to the Patient-AES. Compared to nurses, social workers demonstrated significantly higher levels of patient advocacy engagement (unstandardized β = 6.78, p < .05); no significant differences were found between medical residents and social workers, nor between medical residents and nurses. Hospital site was also not a significantly related to patient advocacy engagement.  Eagerness (unstandardized β = 1.44, p < .001) and Skills (unstandardized β = 0.47, p < .001) were both highly significant predictors of patient advocacy engagement. Belief the Hospital Empowers Patients (unstandardized β = 0.62, p < .05) was also significantly related to patient advocacy engagement. All three of the scales were positively related to patient advocacy engagement, indicating that both individual professional emotional and cognitive characteristics along with the hospital characteristic of empowering patients are relevant. Results are displayed in Table 3.
While all of the predictor scales were significantly related to the Patient-AES in bivariate analyses, the regression analysis demonstrated that Ethical Commitment, Tangible Support, Organizational Receptivity, and the Belief Colleagues Engage were not in themselves independent predictors of patient advocacy engagement. The significant bivariate correlations found between these predictor scales and patient advocacy engagement seems largely explained by their strong interrelationships with both skills and the extent the hospital is perceived as empowering patients. Social workers are trained in the empowerment of vulnerable populations and these skills are likely to flourish in a perceived hospital environment that both values empowerment and in which the network in the hospital of other professional groups are also extensively engaged with advocacy. Nurses, on the other hand, are trained more to attend to the individual biomedical needs of the patient and are not so oriented in the social network of professions in the hospital and the power of the patient in the hospital context.

DISCUSSION
This study sought to understand the predictors of health professionals' engagement in policy advocacy in acute-care hospitals. The analysis demonstrated that all seven predictor scales were correlated with patient advocacy engagement, but only Skills, Eagerness, and the Belief the Hospital Empowers Patients can rightly be conceptualized as independent predictors of patient advocacy engagement. It can be hypothesized that the extent that other professionals engage in advocacy may be the result of the commitment of the hospital to patient empowerment. This suggests, perhaps paradoxically, that those hospitals that empower patients also create a climate where professionals will increase their engagement with patients. That is, professionals will engage in advocacy more extensively when they perceive that the hospital they work in values patient empowerment. This may also explain the strong effect on patient advocacy from social workers than for nurses. Our findings demonstrate that highly rating one's patient advocacy skills is related to engaging in significantly higher levels of patient advocacy. This may be because advocacy requires a specific set of skills that are different than those employed in traditional clinical practice (Jansson, 2011). In our scale that measures the extent health professionals possess requisite skills, for example, we included such skills as "negotiating and bargaining on behalf of specific patients," "resolving conflicts that arise during patient advocacy," "using intermediaries to facilitate patient advocacy," "calling specific unresolved problems to the attention of the attending or consulting physician," and "coaching patients to advocate for themselves." Our findings suggest the need for advocacy training that equips health professionals with these skills. This may include observing seasoned professionals engaging in patient advocacy and placing trainees in role-playing exercises that require them to employ specific skills.
Our findings demonstrate that health professionals who report being eager to engage in higher levels of patient advocacy in the future are more likely to engage in patient advocacy. Professionals who have high levels of eagerness may have particularly high levels of empathy with patients with unresolved problems-empathy that motivates them to greater advocacy engagement. Professionals who have high levels of eagerness may be more aware than other professionals that patients possess many of the 26 kinds of unresolved problems-awareness that increases their eagerness to provide higher levels of patient advocacy. High levels of eagerness may help health professionals, in turn, surmount barriers to advocacy that exist in hospitals that discourage the provision of advocacy, such as the lack of time stemming from their work loads, as well as difficulties that they sometimes encounter from patients, administrators, and other health professionals when they engage in advocacy (Donaldson, 2007;Weiss-Gal & Gal, 2009). This suggests the need to develop advocacy training that increases health professionals' levels of eagerness, possibly by increasing empathy, awareness, and willingness to surmount barriers to enhance patients' well-being.
Our findings demonstrate that health professionals who believe that the hospital empowers patients to advocate for themselves-such as by referring them to support groups, helping them get second opinions, and connecting them to other patients-are more likely to engage in advocacy. In other words, they are more motivated to engage in patient advocacy themselves if they believe that a culture of patient empowerment exists in their hospital. They may view empowering patients as a particularly high level of patient advocacy because it requires substantial interaction with patients to motivate them and to teach them how to advocate for themselves. This finding suggests that advocacy training that motivates considerable numbers of professionals in specific settings to empower patients may induce more reluctant colleagues to empower patients themselves-and even to discuss empowerment strategies and accomplishments with them. Implicit in this finding is the need for hospital-wide advocacy training that results in large numbers of professionals in each of the three groups who actively empower their patients, thereby shifting the culture of the institution to one focused on patient empowerment.
Our findings demonstrate that younger health professionals are more likely to engage in patient advocacy than older professionals. This may be attributable to major changes that have occurred in the American medical system in the past two decades. Considerable health literature has recently focused on the needs of members of vulnerable populations, such as persons of color, disabled persons, members of the lesbian, gay, bisexual, and transgender (LGBT) community, veterans, the elderly, and individuals suffering from mental illness (Jansson, 2011). The term "patient-centered care" has emerged in considerable health literature (Bergeson & Dean, 2006;Kupfer & Bond, 2012). The organization of medical care has changed markedly in the United States as many physicians have become employees of hospitals and as hospitals have increasingly been subject to evaluation of their care by national authorities and by ranking agencies (Committee on Quality of Health Care in America, 2001). In this environment, physicians may increasingly view themselves as part of a health care team rather than solo practitioners-and increasingly invite feedback from nurses and social workers (Mayer, Cates, Mastorovich, & Royalty, 1998). Regardless of profession, the finding that younger professionals are more likely than their older colleagues to engage in patient advocacy may suggest the need for advocacy training across age groups with a style of teaching that invites younger professionals to share their views with more experienced colleagues.
Our findings demonstrate that social workers are more likely than nurses to engage in patient advocacy, but not more likely than medical residents. We think this may occur because social workers do not focus on physiological factors in their training, but in helping patients from a biopsychosocial or person-in-environment perspective (Itzhaky et al., 2004). In addition, social workers are somewhat outside the traditional doctor-nurse hierarchy in which nurses are tasked with carrying out physicians' orders. Since social workers are independent of this chain of command, their "outsider" status may afford them the freedom to engage in greater levels of advocacy. This finding suggests the need for multiprofessional advocacy training so that social workers can impart their proclivities toward and competencies in patient advocacy with other professionals.
Further research is needed to understand the roles of factors such as health professionals' ethical commitment, tangible job supports, the hospitals' organizational receptivity to advocacy, and perceptions of the extent colleagues engage in patient advocacy in influencing their patient advocacy engagement. Each of these scales was statistically associated with patient advocacy engagement in bivariate analyses, yet none emerged as significantly associated with patient advocacy engagement in the multiple regression. Further research should explore their mediating or moderating effects.
Literature suggests that projects to train professionals in advocacy exist, but have yet to be subjected to rigorous evaluation (Bandiera, 2003;Chamberlain et al., 2013;Klein & Vaughn, 2010;Seal, 2007;Wright et al., 2005). We believe that findings from the current study can facilitate the development of patient advocacy training programs in nursing, social work, and medical schools and departments so that their graduates are already prepared to engage in patient advocacy when they enter their professional settings. The Patient-AES (Jansson et al., 2015) may serve as a pre-and posttest measure to evaluate the effectiveness of patient advocacy training.

LIMITATIONS
The study's findings should be considered within the context of its limitations. The cross-sectional design precludes our ability to draw causal inferences. Significant associations between the independent variables and the dependent variable should be understood as correlational but not causal in nature. 576 B. S. Jansson et al.
The eight participating acute-care hospitals-all based in Los Angeles County-may not be representative of hospitals in other regions, thus limiting the study's generalizability. Our participant recruitment strategy yielded a low response rate overall, and differential response rates by hospital and by profession. We surmise this is related to acute-care health professionals' heavy workload. Other researchers have reported similarly low response rates (Dodd et al., 2004). Moreover, participants self-selected into the study, and those who chose to participate may have been more inclined toward the topic of patient advocacy. Their responses may thus not be representative of persons who chose not to participate. These limitations have implications for the study's generalizability to members of these health professions. Moreover, the study relied on respondents' self-reports rather than other sources of data, such as medical records and patients' reports. Participants' confidentiality was protected by this project's recruitment strategy, but not their anonymity because they were recruited by name from hospital rosters. Thus, their responses may have been biased, limiting the study's internal validity.

CONCLUSION
This study has advanced the science of patient advocacy engagement in several ways. The results of bivariate analyses demonstrate that all seven predictor scales considered in this study are significantly associated with patient advocacy engagement, as measured by the Patient-AES. Results of a multiple regression identified that only three of them predict patient advocacy engagement by frontline health professionals when controlling for all other scales and several sociodemographic variables: Skills, Eagerness, and Belief the Hospital Empowers Patients. Data also suggest that younger health professionals are more likely to advocate for patients than older professionals, and that social workers are more likely than nurses to engage in patient advocacy.
The bivariate and multiple regression results provide useful information to the heads of nursing, social work, and medical residency programs and units in acute-care hospitals that can be used to develop patient advocacy training programs. The scales developed in this project can be used as pre-and posttest measures for training programs to reveal if advocacy training programs actually increase levels of patient advocacy engagement, as well as patient advocacy skills, eagerness, and the extent the hospital empowers patients. It is reasonable to hypothesize that higher levels of patient advocacy will, in turn, increase patients' satisfaction with services and well-being.
These findings suggest that advocacy training that focuses on these factors may effectively increase the extent health professionals engage in patient advocacy. We have identified implications of our findings for the content of advocacy training. These include cross-generational training, cross-disciplinary training, and specific strategies to enhance health professionals' eagerness and their patient advocacy skills, as well as facilitating a culture of patient empowerment through hospital-wide training.