Physicians’ Perceptions of Volunteer Service at Safety-Net Clinics
Perm J 2017;21:16-003
Background: Volunteer physicians are crucial for the operation of safety-net clinics, which provide medical care for uninsured and underinsured populations. Thus, identifying ways to maximize the number of physicians volunteering at such clinics is an important goal.
The poor state of medical care for the uninsured and underinsured population in the US is a continuing challenge. The full rollout of the Patient Protection and Affordable Care Act in 20141 led to a sharp decline in the nationwide uninsured rate in the US, from 18% in the third quarter of 2013 to 11.4% in the second quarter of 2015.2 More than 36 million people continue to be uninsured, however, and many have no access to government-sponsored health programs like Medicaid and/or to Federally Qualified Health Clinics.3,4 This group faces numerous health challenges, including poor access to quality health care and increasing rates of chronic, long-term medical conditions.5-8 These medically uninsured and underserved people, many of whom are undocumented, are the clientele for whom safety-net clinics are designed. Safety-net clinics often operate on small budgets and thus often rely on volunteer physicians.4,9,10 Despite the ongoing need for safety-net clinics, the number of physicians volunteering at these clinics is declining.10-12 A nationwide study of volunteer physicians in community clinics in 2000 revealed many barriers to volunteering at safety-net clinics, including administrative burdens; a lack of supplies, equipment, and medical follow-up at the clinics; a lack of social services for patients; patient noncompliance; insufficient time to volunteer; and concerns about malpractice insurance.13-15 The resulting deficiency in volunteer physicians means that recruitment and retention of volunteers is a pressing concern in such clinics. The problems above are not unique to volunteers in safety-net clinics. Physicians employed directly by community clinics experience many of the same challenges facing volunteers, leading to similar difficulties in recruitment and retention.16-21
Many studies have focused on how to recruit and retain employed physicians in community clinics.16,18-21 One way is to tailor such programs to the physicians’ motivations for working with the underserved. Many clinicians working in community clinics are specifically attracted to the missions of such clinics.16,18-20,22 Curlin et al23 and Cole et al18 found that the physicians’ faith/religion was a common motivator for working with underserved populations at community clinics, but that aspects of the physicians’ personalities (such as the desire to “make a difference”) may be of more fundamental importance. Stevenson et al24 recorded physicians stating that working in community clinics “was the right thing to do” and a source of significant satisfaction. Li et al25 found that physicians working in community clinics tended to state that such work aligned with their overall values and sense of social justice. Exposure to people from such underserved populations while growing up or during medical training appeared to help physicians avoid developing prejudices toward these populations. This led to an overall attraction to the mission of community clinics. Appealing to these values may be very helpful in physician recruitment to community clinic service. Li et al25 also determined that group support and being part of a team were key ways that physicians supported each other and avoided burnout. More structured hours, so that working did not intrude upon personal time, also appeared to be helpful.
One useful theoretical framework for examining the motivation for volunteerism in physicians is that of Clary et al.26 Their work determined six categories of functions for volunteer activity (Table 1). These functions range from providing tangible career and social benefits to manifesting more pure humanitarianism. Some research24,27 suggests that volunteering tends to serve a prosocial role, in that it benefits others but not at the expense of oneself. Prior research using focus groups25 and interviews23 of physicians employed at community clinics have investigated the reasons behind prosocial career choices of physicians, but studies focusing on volunteer physicians are less common.
To serve the growing population of medically underserved individuals, it is imperative to increase the number of physicians volunteering at safety-net clinics.28,29 Toward this goal, in this work we examine the perceptions, motivations, functions, and barriers associated with volunteering in safety-net clinics in a population of physicians belonging to the Southern California Permanente Medical Group (SCPMG) in Southern California. Our results may have implications for increasing physician volunteerism as well as for illuminating the motivations and functions for volunteering.
This study uses a mixed-method approach, comprising a survey instrument with Likert-scale responses and additional invited written responses as well as subsequent face-to-face interviews with a semistructured format.
Survey participants and location
A total of 40 online surveys were distributed to current and past participants in a physician volunteer program supported by SCPMG and its affiliated large health care organization (Kaiser Foundation Health Plan) in Southern California. The physicians volunteer at 3 non-federally funded safety-net clinics and one Federally Qualified Health Clinic that provide both primary and specialty care to underserved and uninsured populations in the area. These clinics are located in an area identified as a Health Professional Shortage Area and a Medically Underserved Area by the US Department of Health and Human Services.30 This region has also been assigned a high Community Need Index, which measures the number of barriers to care that the population experiences, such as unemployment and inability to acquire insurance.31 It is also an area with a recent significant increase in chronic medical conditions, such as diabetes and hypertension.31 The geographic area in this study exemplifies the “inverse care law”: “The availability of good medical care tends to vary inversely with the need for it in the population served.”32
In the physicians’ Medical Group, physicians are given various options for the professional enrichment/service aspect of their jobs. One option is volunteering during discretionary time (time during regular work hours devoted to physician education and personal enrichment) at safety-net clinics in the area. Participating physicians typically volunteer one half-day per month, and some add personal volunteer hours on evenings and weekends. This program contrasts with volunteer programs in prior literature,28,29 in which volunteer physicians were not compensated for their time at all. Another key component of the program in the current study is that the sponsoring Kaiser Foundation Health Plan assumes responsibility for malpractice liability for all volunteer situations.
Survey questionnaire development
The survey questions were designed to explore issues relevant to physician volunteers serving in safety-net clinics. The survey included demographic questions, 9 structured attitudinal items rated on a 5-point Likert scale (strongly disagree, disagree, neither disagree nor agree, agree, strongly agree), and the opportunity to provide written comments for each Likert block, for a total of 49 items. Two split-half reliability assessments supported the reliability of the survey (tb 0.787 and 0.907, respectively; p < 0.0005). A Chronbach alpha further demonstrated survey reliability for 8 of the 9 Likert blocks (a > 0.7).
After approval from the University of Phoenix institutional review board, the survey was administered via SurveyMonkey (SurveyMonkey, Palo Alto, CA). A request to participate in the survey, including a link, was e-mailed to all 40 physicians in the volunteer program. The first page of the survey handled informed consent, and completion of the survey demonstrated implied consent. Responses were recorded anonymously; neither e-mail nor Internet Protocol addresses were collected. Thirty-one physicians completed the survey, with most doing so between October 2014 and November 2014.
Interview question development
Follow-up interview questions were derived from an analysis of the survey to further elucidate perceptions underlying the survey responses. The interview schedule addressed personal motivation, personal meaning of volunteer work in the program, and personal and professional challenges to volunteering.
After the conclusion of the survey, all participants received a letter inviting them to participate in a confidential interview with the first author. Eight of the 40 physicians in the volunteer program agreed to the face-to-face interviews; all 8 confirmed they had also taken the survey. The interviews took place March 2015 to May 2015. The interviewees signed consent forms to be interviewed and recorded; the first author (LM) recorded and transcribed all interviews.
The first author (LM) analyzed interview transcripts with a general inductive approach using qualitative content analysis. The analysis of questions concerning personal motivation and personal meaning highlighted several themes that aligned with the functional categories of volunteerism proposed by Clary et al.26 In addition, inductive codes were developed for the questions concerning personal and professional challenges. We recorded the number of physicians who cited each theme or function, as well as the number of times a certain theme or function was cited across all physicians. The patterns of these two sets of numbers were qualitatively quite similar to each other. Data were coded using NVivo software, version 10.2 (QSR International, Melbourne, Australia).
Thirty-one of the volunteer physicians (78%) completed the online survey (demographic characteristics in Table 2). Most respondents were women; they were predominantly Asian/Pacific Islander or white; more than 80% identified with some religion; the median age was 49; and more than 60% had dependent children living in their households. There was an equal distribution of primary care physicians and specialists. Approximately 74% were active volunteers who had volunteered an average of 10 times. Of 31 surveyed physicians, 5 volunteered at least partially on weekends and 4 volunteered at least partially at night (periods for which their volunteering hours would not be compensated).
The survey produced remarkably uniform results for perceptions of conditions at the safety-net clinics, with positive views about volunteering. More than 75% agreed or strongly agreed with a variety of positive statements concerning perceptions of conditions at their safety-net clinics, indicating a high degree of trust in clinic leadership, high levels of interactions and engagement with clinic staff, and feeling like a part of the group at the clinic. No more than 17% either disagreed or strongly disagreed with any of these statements.
Survey results also revealed information about the personal motivations of physicians in the program. All of the physicians agreed or strongly agreed that volunteering gives them a sense of personal satisfaction, and 96% of them agreed or strongly agreed that they feel positive about their volunteer experience. Personal faith also played a part in personal motivation, with 81% agreeing or strongly agreeing that volunteering is aligned with their faith (note that almost 84% of respondents identified with a religion). Furthermore, 91% of the physicians agreed or strongly agreed that they plan to continue volunteering in the physician volunteer program, and 80% would recommend the program to others.
Consistent with positive feelings about volunteering and agreement about positive motivations was a corresponding rejection of statements about personal and professional challenges to volunteering. Few of the challenges garnered more than 50% “agree” or “strongly agree” responses. In contrast to the uniformity of responses to positive statements, there was a diversity of opinions about specific challenges to volunteer service. The standout personal challenge was that volunteering takes too much personal time, to which 42% of the participants agreed or strongly agreed. The most commonly cited professional challenges were limited supplies at the clinic (58% agreed or strongly agreed), subpar equipment at the clinic (55% agreed or strongly agreed), and that the patients’ social, transportation, and financial challenges impacted medical care (77% agreed or strongly agreed).
In the survey, 4 open-ended questions (associated with Likert blocks) allowed respondents to volunteer written comments. Of 13 comments recorded, 5 were related to motivations for volunteering. Examples include “I find it refreshing to occasionally get away from my normal clinic work setting and offer patients care in a different environment,” and “I see many unusual conditions that improve my skills and understanding of disease pathophysiology.” Eight of the comments were related to professional (organizational and resource-related) challenges to volunteering, such as “In my opinion, the clinic lacks good leadership,” and “Some patients will have needs that I will not be able to provide much for without additional resources.” Comments about personal challenges included “I stopped volunteering because I don’t have time. My daily clinic is more than enough.” These comments aided in the development of questions for the interview section of the study.
The interview portion of the study had a 20% response rate. Our interview participants consisted of 5 men and 3 women, with an even split between specialists and primary care physicians (Table 3). All 8 interview participants had previously taken the survey.
Personal meaning and motivation for volunteer service
The Volunteer Functions Inventory26 was used to interpret both the answers to posed questions and the spontaneous statements of the interview participants. All questions were analyzed to tabulate the number of citations of specific functions, and most of the citations were found in two questions: 1) What motivated you to be a volunteer in this program? 2) What does your volunteer work mean to you personally?
The most cited function was Values, defined by Clary et al26 as “The individual volunteers in order to express or act on important values like humanitarianism” (Table 1). All 8 interviewed physicians cited this function, which had 38 separate survey citations (45% of all function citations). The next most commonly cited function was Protective: “The individual uses volunteering to reduce negative feelings, such as guilt, or to address personal problems.” Six interviewees cited this function, with 20 separate citations across all physicians (24% of function citations). Less commonly cited was the Enhancement function: “One can grow and develop psychologically through volunteer activities.” Six interviewees also cited this function, but it had only 15 citations across all physicians (18% of the function citations). A smaller number of physicians cited the 3 remaining functional categories (Understanding, Social, and Career), and each of these functions comprised a significantly smaller percentage (< 8%) of the citations across all physicians.
The two most commonly cited functions merit additional analysis. The Values function, in which the physicians are volunteering because such behavior is aligned with their personal values (such as humanitarianism), was defined broadly during the interviews, including statements expressing that volunteering is “doing the right thing,” recognizing that “people need help,” that volunteer service is “part of my faith,” and that prior experience is an influence (eg, “I have always volunteered,” “I want to give back,” and “it was part of my training”). Personal satisfaction associated with the Values function was seen in a wide variety of statements, including these:
“It actually makes me feel like I’m helping someone, and that in itself is really kind of gratifying to me, and also within my faith. It is not required, but it is highly encouraged to help your neighbor.”
Some physicians reported that it was in their moral code to “give back” if they had been recipients of generosity in the past:
“I just want to give back. I got so much, so I thought I should give back something. So, it’s just, you know, my background is different. … I got people who come out to help me, volunteer people.”
Many physicians stated that volunteering provided benefits that were perceived by the physicians as beyond those that they received from their home clinics (another Values item):
“It kind of brings an extra dimension to what work itself is, like, it takes my skills and makes it more than something that I use to earn money.”
Physicians also expressed statements that demonstrated how their volunteer service served the Protective function, which is related to preventing or alleviating negative feelings in one’s life. In many cases, physicians stated that volunteering served as a break from their normal duties:
“No, the volunteering does not add to the burnout. I’ll be honest with you, if I feel burnout, I would not be volunteering. However, it is a positive feedback, it actually helps to alleviate my burden with my regular work.”
Challenges to volunteering
In addition to illuminating the various functions of volunteering, the interviews also investigated personal (Table 4) and professional (Table 5) challenges associated with volunteering. By far the most common personal challenge cited was time. Physicians expressed statements such as “but you know, we are so busy, with family, with life, with career.” Some volunteers noted that often they would rather do something more relaxing than volunteering, such as reading papers, or catching up on professional paperwork during their discretionary time. Many physicians also cited time concerns as a primary reason for feelings of burnout in their regular jobs. Several cited the role of the electronic medical record (EMR) in taking time from patient interactions in their regular clinics (eg, “The computer has made you a slave”). In contrast, many physicians stated that their volunteer time was not as dominated by paperwork and an overwhelming array of responsibilities as their standard duties; as noted in the Protective function analysis above, they found that volunteering served as relief from the stresses of their normal job.
The interviewees cited a large and varied array of professional challenges to volunteering (Table 5). The most commonly cited professional challenge was the organization of the safety-net clinic; all 8 physicians cited this challenge, which accounted for 31% of all the professional challenges cited. Physicians noted that the number of clinic staff was often too small to be effective, the staff lacked adequate training, and follow-up or tracking of patients was inadequate. Additionally, scheduling was sometimes ineffective, with volunteer physicians sitting around with no patients to treat at times, and too many patients at other times (“They called me and said, ‘Can you see 6 patients?’ and I said, ‘Can you get 6 patients in the room?’”).
The second most commonly cited professional challenge was problems with supplies at the clinic; 7 physicians cited this challenge, which accounted for 21% of the total. These problems included a lack of supplies for patient care, as well as a lack of EMR and computers. The third most commonly cited professional challenge was the organization at the physician’s home medical organization, which administers the volunteer program. This challenge was cited by all 8 physicians and was 19% of the challenge citations. In particular, physicians reported a need for more flexible scheduling from the home medical organization, more incentives to volunteer, and more mentoring for the volunteers. Challenges cited at a smaller rate included problems with the patient population (including noncompliance with medical advice and a lack of money and insurance), and a lack of mental health services, social work, and other specialized procedures at clinics. Two physicians initially stated that there were very few barriers to their volunteering, although they later did cite some challenges.
The results of the combined survey and interviews describe a physician population that is highly motivated and possesses quite positive perceptions about volunteering in safety-net clinics. These results concur with previous studies that show positive perceptions towards volunteering by US physicians,10,15 as well a commitment of physicians to social justice and the mission of helping underserved populations.18,23-25
The physicians in this study reported feeling appreciated at their safety-net clinics, which has been shown to be associated with resilience in community clinics.9,33 Survey participants indicated that they felt like part of the group when volunteering and interacting with other providers at the clinic. This finding may be an important indicator of physician wellness, which has been studied in physicians’ regular clinics34 and has been cited as a key factor for other providers who work in community health.25 Feeling like part of the group is consistent with the Social volunteering function of Clary et al.26 However, none of the eight interviewed physicians mentioned the Social function in answering questions about personal meaning and motivation. This seemingly contradictory result between the survey and the interviews may imply that although the physicians in our study recognize that the social aspect of volunteering is positive, it is not forefront in their concerns.
Our surveyed physicians agreed with a wide array of potential motivating factors for volunteering, from personal satisfaction to positive statements about community health (including an admiration for the underserved population) and volunteering being an asset to their professional careers. This result is consistent with that of Stevenson et al,24 who showed that admiration for the patient population can help sustain physicians working in challenging conditions.
In contrast to the relative agreement in the survey about many different positive factors, there was no such consensus on challenges to volunteering (with the exception of time concerns). As noted above, only 3 of 21 potential challenges listed in the survey produced a majority of agree or strongly agree responses. Indeed, the follow-up interviews identified some physicians who claimed that there were no significant barriers to their volunteering. Previous surveys have cited professional challenges to volunteering15 similar to those in our survey. However, in agreement with qualitative research by Stevenson et al,24 physicians in the current study downplayed these challenges in their personal interviews.
Both the survey results and the results of the interviews indicate that the primary motivations and benefits of volunteering for the physicians relate to the Values and Protective functions of Clary et al.26 The relationship between volunteering and Values is quite logical; in both the survey and the interviews, physicians indicated that volunteering was part of their moral code and often aligned with their faith. An emphasis on personal values and faith is consistent with the results of focus group25 and interview23 studies with physicians employed at community health clinics. In particular, Li et al25 found that employed physicians working in community clinics cited many of the same motivations and sources of satisfaction as our current study on volunteer physicians. Many physicians in our study, like those in the work of Li et al,25 also had previous experience with underserved populations in their medical training, and had previous experience with volunteering. It should be noted that many of the Value function examples cited by the physicians correspond to prosocial behaviors (ie, behaviors that benefit others, but not at the expense of the self).24,27 Previous research has suggested that an emphasis on the “prosocial, rather than the purely altruistic” aspect of volunteerism may help recruit and retain volunteers.27
Perhaps surprising is the idea, highly cited in the interviews, that volunteering serves the Protective purpose of Clary et al26—specifically as protection from feelings of burnout. Prior studies of physician well-being34 indicate that the parts of physicians’ jobs that are most frustrating and tend to lead to burnout are nonmedical, including EMR, documentation, and dealing with regulations. Another large factor in burnout is the need to deal long-term with patients with complicated medical problems or severe illnesses.35 The present results indicate that volunteering may help alleviate both of these causes of burnout in physicians. Because the volunteer physicians are part-time at the safety-net clinics, they are not responsible for many of the nonmedical duties, so they can focus more on patients (although some do complain about the lack of EMR at the clinics). In prior work, interviews with physicians in Australia revealed that relief from such organizational problems and control over the number of hours worked led to a greater sense of well-being.24 Because patients at the safety-net clinics are treated on an outpatient basis, the volunteer physicians in our study do not typically have the “heart sink” of long-term follow-up with patients with complicated medical issues.24 Rather than being difficult owing to spillover of stress or burnout from the physicians’ primary clinic, volunteering can serve as an effective relief from work-related pressures for these physicians. Promoting the idea that volunteerism may provide relief or prevention from burnout may also prove useful in recruitment and retention of volunteer physicians.
An important personal challenge to volunteering illuminated by both the survey results and the interviews is time—the intrusion of volunteering on one’s personal time and one’s time at work to perform nonmedical tasks. Lack of time was one of the most highly agreed challenges on the survey, and the single most highly cited challenge in the interviews. This result is in agreement with prior work with physicians at community clinics.18 Our results, consistent with those of Li et al,25 imply that better time management in volunteer programs, including the ability to set one’s own hours, may be a key improvement for better recruitment and retention in such programs. Our volunteers, however, perhaps owing to the smaller number of hours worked and the fact that much of this work was part of their discretionary time, do not consider time concerns to render their work unsustainable, in contrast to some of the employed community clinic physicians in Cole et al.18
The current study has some limitations that must be noted. Although the survey sample included 75% of the physicians in the volunteer program, those physicians still may not be representative of the physicians of the entire SCPMG, or of physicians in the area as a whole. Because our subjects by definition have already chosen to participate in the volunteer program, they may differ in important ways from physicians who have not chosen to participate; as an obvious example, they may be more highly motivated to volunteer.33 Also, it is possible that those who volunteer primarily on their personal time may differ in significant ways from those who volunteer during their discretionary (paid) time, but our data do not allow us to break these groups apart from each other. The follow-up interview pool is even smaller than our survey pool. Those who volunteered to be interviewed may differ in systematic ways from those who took the survey but chose not to be interviewed. One may speculate that those with more negative perceptions of the volunteer program may not be comfortable being interviewed, even with the assurance of anonymity. Those who chose to be interviewed also may be more resilient to the challenges of volunteering. Finally, volunteer physicians working in other geographic areas may serve different patient populations, and may face different challenges, so our results should not be overgeneralized.
The results of this study may have implications for the recruitment and retention of physicians at safety-net clinics and for the delivery of medical care to underserved populations in Southern California and beyond. This is a burning issue that has been studied in the context of employed19-21 and volunteer28,29 physicians. As noted by Clary and Snyder36 and documented by Ahmed and Maurana28 and Shuman et al,29 recruitment and retention should match the functions that volunteering serves for the volunteer population. An obvious first-order point is that the physicians in the current study have very positive opinions overall of their volunteer experience; this positivity could be made much clearer to potential volunteers for recruitment. In particular, appeals to physicians’ sense of values (and perhaps targeted appeals to physicians of different faiths22) could draw more physicians to such programs. An additional point that could be made is that volunteering may serve as a crucial “escape hatch” from the stresses of their regular jobs—in other words, volunteering could have a valuable function in burnout prevention. Conversely, given the lack of emphasis that the physicians placed on technical skill and career advancement, stressing these benefits may not be as effective. Direct personal contact may be key in emphasizing the benefits of volunteering to potential new volunteers.28 It may also be possible to target for recruitment physicians whose personal attributes (eg, faith, prosocial tendencies, a sense of social justice) lend themselves to deriving satisfaction from work with underserved people at community clinics. As suggested by Li et al25 and Curlin et al,23 it may be possible to teach such values in medical school, and specifically nurture candidates with these traits and steer them toward participation in community clinics.
Once physicians have joined a safety-net clinic or volunteer program, this study suggests that fostering a sense of being part of the group in the clinic may be a subtle but important factor in increasing physician satisfaction and improving retention; proper mentoring may also be important. Organization at the clinics in our study may need some improvement, although many of the issues at the clinics (including lack of EMR, supplies, pharmaceuticals, and transportation and follow-up for patients) are not easily solvable without significant additional funding.4 In addition, this study indicates the importance of having regular work time set aside for volunteering. It also may be quite helpful to find ways to keep volunteering from intruding on personal time, as well as to have more effective and efficient scheduling of patients. It is hoped that studies such as this may help organizers structure their volunteer programs to better meet the needs of volunteer physicians, and thus more effectively attract and retain participants. Even with the advent of nationwide health care reform in recent years, the need for safety-net clinics targeted to underserved populations is not diminishing, and may in fact be growing. More work is clearly necessary to recruit physicians to volunteer at such facilities.
The author(s) have no conflicts of interest to disclose.
This work was supported by the School of Advanced Studies, University of Phoenix, AZ. The authors gratefully acknowledge local and regional Kaiser Permanente and Southern California Permanente Medical Group leaders for their support, and give special thanks to the many SCPMG physician volunteers who give so much of themselves in their volunteer work and in helping with this research.
Mary Corrado, ELS, provided editorial assistance.
How to Cite this Article
McGeehan L, Takehara MA, Daroszewski E. Physicians’ perceptions of volunteer service at safety-net clinics. Perm J 2017;21:16-003. DOI: https://
1. Bureau of Primary Health Care, Health Resources and Services Administration. HRSA health center program [Internet]. Rockville, MD: Health Resources and Services Administration; 2014 [cited 2016 Mar 23]. Available from: http://bphc.hrsa.gov/about/healthcenterfactsheet.pdf.