Improvements in medical school wellness and career counseling: A comparison of one-on-one advising to an Advisory College Program

Abstract

Background: Medical students have unmet needs in the areas of career and wellness advising.

Aims: The goal of this study is to describe the development of an Advisory College Program (ACP) and assess its effectiveness compared to a traditional one-on-one faculty advisor system.

Methods: The ACP, consisting of four colleges co-led by Advisory College Directors and supported by key Faculty, was developed to provide structured career and wellness advising. The authors compared the ACP to the former Faculty Advisory Program (FAP) using two parallel questionnaires.

Results: Surveys were completed by 74% of first-year students, 60% of second-year students, and 88% of third-year students. Survey data demonstrated a significant increase in the number of students who could identify their advisor, the frequency of student–advisor contacts, and the perceived accessibility of advisors in the ACP compared to the FAP. While an ordinal logistic regression model did not demonstrate a significant effect of the new advising system on overall satisfaction, univariate analysis demonstrated a significant increase in student satisfaction with wellness and career counseling.

Conclusions: The ACP was more effective in promoting student wellness and career counseling than the traditional one-on-one faculty advisor system. Similar college-based programs may be beneficial to students at other medical school programs.

Introduction

As medical students navigate through their preclinical and clinical years, almost all will encounter some degree of difficulty with personal wellness or career development. Many medical schools have attempted to cultivate professional and career development among medical students by assigning faculty advisors to individual students (Mann 1992; Rose et al. 2005). Studies have demonstrated that mentoring relationships play a role in career advancement (Jackson et al. 2003) and increase professional satisfaction and productivity (Sirridge 1985; Ramanan et al. 2002). While the impact of medical student mentorship on career success is unknown, students report that mentors commonly provide career support (Aagaard & Hauer 2003) resulting in overall increased student satisfaction (Coates et al. 2008).

Defined medical student advising systems can also address the high rates of professional distress present in medical schools today. A systematic review demonstrated significantly higher levels of depression and anxiety among the US and Canadian medical students as compared to age-matched controls and the general population (Dyrbye et al. 2006b), while Tyssen et al. (2001) reported a high prevalence of suicidal ideation among medical students. In a 2006 multicenter study, a striking 45% of students reported burnout (emotional exhaustion, depersonalization, and low sense of personal accomplishment), and the rates of depersonalization increased during the medical school (Dyrbye et al. 2006a). Rates of depression in these students were highest in the second year of medical school (69.6% of students surveyed), suggesting that early intervention is needed to address these symptoms of professional distress (Dyrbye et al. 2006a).

While the above studies support the need for the creation of effective student wellness and career advising systems, multiple studies have discussed the difficulty of constructing successful formal programs (Mann 1992; Igartua 1997; Deangelis 2004) prompting medical schools to create novel advising systems (Puckett et al. 1989; Kalet et al. 2002; Murr et al. 2002; Scheckler et al. 2004; Goldstein et al. 2005; Macaulay et al. 2007; Stewart et al. 2007; Coates et al. 2008). We developed a fundamentally different approach to student advising in view of the need for our medical students to find mentors and in recognition of the perceived lack of success of current advising systems. Rather than relying on volunteer faculty and loosely defined roles, we created an Advisory College Program (ACP) whose aim is to support a core group of interested faculty engaging their students in structured activities. We hypothesized that the ACP would be more effective than a one-on-one Faculty Advisory Program (FAP) for promoting medical student wellness and providing career counseling.

Table

Practice points

Medical students have unmet needs in the areas of wellness and career counseling. Current advising systems have been unable to demonstrate effectiveness in promoting student wellness and career counseling. The Advisory College System was associated with an increase in students’ abilities to identify their advisors, the number of student–faculty interactions, and the perceived effectiveness of the new advising system. We anticipate that improved student advising techniques may help address medical student burnout and depression as well as address career counseling needs.

Methods

The current structure of the ACP was developed by a group of interested students and the Associate Dean for Student Affairs and was implemented in July 2007. It replaced the traditional FAP present in most medical schools today, in which each medical student is paired with a volunteer faculty member who serves as the student's advisor throughout their undergraduate career. In contrast, our ACP consists of four advisory colleges each co-led by two faculty members called Advisory College Directors who have been nominated by a student selection committee and approved by the College of Medicine Deans. Each advisory college is also supported by approximately six Faculty Affiliate Advisors who represent a broad range of clinical disciplines from across the medical center. The students are equally divided between the colleges and remain in their advisory college throughout their medical school training. Thus, with Vanderbilt's current size, each college has approximately 25 students from each class or 100 students total.

The two primary pillars of the ACP are to promote personal wellness and assist in career development. Scheduled and impromptu meetings between advisors and students provide a forum for targeting individual needs in these areas. Additionally, college-wide activities focus on building community within each college and promoting interaction between students in preclinical and clinical years. Students can therefore be secure in the knowledge that they are part of a family of support. Examples of these activities are outlined in Table 1. Each advisory college receives a stipend to support group activities, and the Advisory College Directors receive salary support. The eight Advisory College Directors and the Associate Dean of Student Affairs meet monthly to plan intercollegiate events, set program goals, and share successful advising strategies. This structure provides oversight and accountability and improves consistency across the colleges.

Table 1.  Annual activities within the Advisory College System, Vanderbilt University School of Medicine

Calendar month	Event	Description	Primary focus
August	Welcoming parties	College-wide social event at beginning of academic year	Wellness (focus on transitions)
September	College cup	Intercollegiate Olympics-style competition	Wellness (physical fitness, team building)
September–November	College-wide social gatherings	Build community within each college	Wellness and career counseling
September–November	Advisory college director meetings	Individual sessions between students and their advisory college directors	Wellness and career counseling
October	Wellness retreat	Day retreat for first- and second-year students	Wellness (coping skills, balance)
January–April	College-wide social gatherings	Build community within each college	Wellness and career counseling
January–April	Advisory college director meetings	Individual sessions between students and their advisory college directors	Wellness and career counseling
March	Career specialty fair	Small-group sessions and career fair	Career counseling
April	Shade Tree Trot, 5 K Run	Intercollegiate charity event and fundraiser	Wellness (college unity)

We developed a study to assess medical students’ perceived effectiveness of the ACP compared to the former one-on-one FAP. The Institutional Review Board at Vanderbilt Medical Center approved this study. We began formative assessment of the program within its first year by developing two parallel questionnaires to address the perceived effectiveness of the system in question and the role of the advisor in promoting wellness and career counseling. We developed these unique questionnaires in order to specifically evaluate student satisfaction and well-being as they related to the goals of the advising programs.

The FAP survey was provided to second- and third-year medical students at the beginning of the academic year 2007–2008 and consisted of 53 items regarding their experiences with the FAP the previous year. The ACP survey was similar to the previous survey but included 12 additional questions related to Affiliate Advisors. It was delivered to all first-year students at the end of the academic year 2007–2008 and was based on their experiences with the ACP. Students were provided with surveys by hand at the end of three different class sessions to increase elective participation in the study. All questionnaires were anonymous.

Data from both sets of questionnaires were collated into a database. We compared parallel questions from the two surveys using a 2-sample t-test of proportions with a continuity correction for binary questions and a χ²-test for questions scored on a 5-point Likert scale. In addition, data were analyzed using an ordinal logistic regression model and ANOVA. For this model, the statement “I am satisfied with the overall advisory system” was used as the response variable. Students were able to respond to this question with five possible answers, including strongly disagree, disagree, neutral, agree, or strongly agree. Since many of our items were not continuous and some of the categories were not properly balanced to assume continuity, we chose to perform a cluster analysis wherein variables relating to the above statement were separated into eight clusters using Spearmen's rho which allowed us to use the ordinality of the measures. Prior to reviewing the data, four different rules were used to make up the eight clusters based on scoring done by the clinicians and students performing this study. Since less than 5% of the data were missing on any of the items, we used simple imputation for the small portion of missing data. Using a proportional odds model with restricted cubic splines of four knots, the likelihood ratio and AIC were used to evaluate which scoring scheme fits the model best. The model was evaluated using a bootstrap of 150 repetitions to see how well it would hold up in future data sets. The corrected Somer's D was 0.82, suggesting that this model would perform relatively well on future data sets. The calibration plot also showed that the model calibrated well.

Results

The ACP survey was completed by 76/103 (73.7%) first-year students. The survey for the FAP was completed by 65/109 (59.6%) second-year students and 93/106 (87.7%) third-year students. Demographic data collected from participating students included gender, race, nationality, research interest, and relationship status. There were no significant differences between the demographics of students in each survey group, and the characteristics of the participating students mirrored the entire medical school class.

The proportion of students who knew their advisors in the ACP was significantly higher than those in the FAP (100% as compared to 78%, respectively; p < 0.001). There was also a significant shift in the number of contacts between students and their assigned advisors following implementation of the ACP (Figure 1; p < 0.001). In the FAP, 56% of the students reported never being contacted by their advisor, and only 5% of the students reported ≥5 contacts. In comparison, 100% of the students in the ACP were contacted by their advisors, and 55% reported that they were contacted ≥5 times. When asked about meaningful contacts with their advisors, 53% of students in the FAP reported no meaningful contacts, while only 25% had >2 meaningful contacts. Conversely, 87% of students in the ACP reported >2 meaningful contacts in the preceding year.

Figure 1. Distribution of the frequency of advisor–student contact in the FAP and the ACP.

In both surveys, over 94% of the students agreed that wellness and career advising was important to them. We analyzed student satisfaction with these two components in each program (Figure 2). Twenty-seven percent of students in the FAP agreed or strongly agreed that they were satisfied with how well their advisor promoted wellness, compared to 72% of students in the ACP (p < 0.001). Similarly, only 22% of students in the Faculty ACP agreed or strongly agreed that they were satisfied with how well their advisor provided career counseling, compared to 43% of students in the ACP (p < 0.001).

Figure 2. Distribution of students’ responses in each advising program (FAP and ACP) for questions: (1) Wellness Advising – “I am satisfied with how well my advisor promotes wellness” and (2) Career Advising – “My advisor has been effective in providing me with career counseling.”

We found a significant (unadjusted) association between the mentoring system and the level of satisfaction for specific components of wellness advising, career advising, and overall student advising (Table 2). In general, a higher proportion of students agreed that the ACP advisors broached topics of personal wellness, such as exercise, sleep, stress, and mental health as compared to advisors in the FAP. Similarly, specific issues related to career counseling, such as creation of professional goals and provision of opportunities for career development were more likely to be addressed by advisors in the ACP. For general questions about advisors, such as approachability, accessibility, and overall influence, there was a significant improvement with the ACP.

Table 2.  Percentage of students who agree or strongly agree with questions related to wellness advising, career advising, and overall advising in the FAP and the ACP

	FAP (agree or strongly agree)	ACP (agree or strongly agree)	Test statistic
Wellness advising questions
My advisor talks to me about healthy sleep habits	8%	55%	= 63.36, p < 0.001
My advisor talks to me about healthy exercise habits	14%	55%	= 43.93, p < 0.001
I would feel comfortable discussing my personal stress with my advisor	24%	62%	= 40.9, p < 0.001
I would feel comfortable discussing my mental health with my advisor	27%	51%	= 31.84, p < 0.001
Career advising questions
My advisor encourages me to develop professional goals	52%	71%	= 47.05, p < 0.001
My advisor provides me opportunities that aid in career development	19%	38%	= 19.96, p < 0.001
My advisor explores specialty options with me	14%	38%	= 29.73, p < 0.001
My advisor explores research opportunities with me	13%	25%	= 20.29, p < 0.001
General advising questions
My advisor is approachable	61%	94%	= 29.87, p < 0.001
My advisor is accessible	57%	90%	= 30.21, p < 0.001
My advisor has had a positive influence on my medical school experience	38%	84%	= 46.58, p < 0.001
I am satisfied with the overall advisory system	26%	68%	= 42.51, p < 0.001

Based on the analysis of the single question “I am satisfied with the overall advisory system,” there was a statistically significant improvement in the ACP compared to the FAP (67% vs. 24%; p < 0.001; Figure 3). Using our eight clusters (Advisors’ Interest, Interaction with Advisors, Career Exploration, Career Advising Values, Wellness Advising, Support from Advisors, Quality of Advisors, and Advisor's Approachability), we sought to determine the relative importance of each of these themes to the significant changes noted in the students’ satisfaction levels. An ordinal logistic regression model to evaluate students’ overall satisfaction found the important predictor theme clusters to be Interaction with Advisors, Support from Advisors, and Quality of Advisors. The remaining five themes as well as our main variable of interest, the Advisory System as a whole, were not statistically significant in relation to students’ overall satisfaction with their current advising system.

Figure 3. Distribution of students’ responses to the question “I am satisfied with the overall advisory system” – “system satisfaction.”

Discussion

Undergraduate medical students have unmet needs in the areas of personal wellness and career development, necessitating a redesign of current student advising systems. Physicians at all stages of training perceive the importance of mentorship (Sambunjak et al. 2007) and while nearly all students desire mentoring relationships (Ricer et al. 1995; Aagaard & Hauer 2003), few of them report having an effective advisor (Aagaard & Hauer 2003). This discrepancy has many causes including lack of contact with clinical faculty during preclinical years, the short duration of classes and rotations, and the need to rely on the volunteer efforts of faculty (Igartua 1997; Hauer et al. 2005; Rose et al. 2005). Since reliance on the spontaneous formation of effective mentoring relationships is inadequate, most schools have developed formal programs to increase student/faculty contact and promote career advising (Mann 1992; Igartua 1997). Unfortunately, systematic reviews of these programs have been unable to demonstrate their effectiveness (Buddeberg-Fischer & Herta 2006; Sambunjak et al. 2007).

In addition to students’ need for mentors to aid in career development, there is growing concern related to the unmet wellness needs of medical students. Medical students have been shown to demonstrate a loss of positive health habits and coping skills (regular exercise and meaningful social interactions) during their first years of medical school (Ball & Bax 2002). At the same time, rates of alcohol use and binge drinking increase during this period (Mangus et al. 1998; Ball & Bax 2002) suggesting that positive coping skills are being replaced with patterns that are markers of potential alcohol abuse and neglect of personal wellness. Student psychological distress may impact academic performance, as there is empirical evidence of an inverse relationship between medical student stress and academic success (Stewart et al. 1999). Finally, it has been demonstrated that there is a statistically significant decline throughout medical school training in student attitudes toward the doctor–patient relationship, relationships with other healthcare workers, and social factors affecting patients (Woloschuk et al. 2004). Researchers have linked this decrease in medical student empathy to increased burnout and distress, whereas high quality of life ratings correlated positively with higher empathy scores (Thomas et al. 2007). These findings provide a strong motivation to create programs to address student wellness issues (Thomas et al. 2007).

In recognition of these deficits in career development and wellness promotion, many programs across the country are adopting new strategies to address the advising needs of medical students. We replaced the traditional one-on-one faculty advisory system with a structured ACP designed to facilitate student–faculty interactions and promote student wellness and career guidance. While other institutions have created similar systems, we believe this is the first program to formally evaluate a college-based advising system in comparison to a more traditional faculty advisory system.

Using an ordinal logistic regression model, we were unable to demonstrate a statistically significant effect of our new system on overall satisfaction with student advising. However, the univariate analysis did show that all the individual items we evaluated were highly significant (Table 2) suggesting that items within the predetermined clusters did play a significant role even though the clusters themselves did not. We also feel that we have made significant improvements in the overall advising system that are likely to impact satisfaction with student advising. For example, limited contact with faculty has been noted to be one of the barriers preventing students from establishing adequate mentoring relationships (Mann 1992; Igartua 1997). We report a significant increase in the number of students who could identify their advisor, the frequency of meaningful student–advisor contacts, and the perceived approachability and accessibility of advisors in the ACP compared to the FAP. We believe these improvements will lead to greater satisfaction with the advising system and promote more effective mentoring relationships.

Our survey data also suggested an association between implementation of the ACP and students’ perception of effective career guidance. Specifically, we demonstrated an increase in students’ perception of their advisors’ abilities to help them develop professional goals, explore research opportunities, and discuss specialty options. Students not only noted that their advisors demonstrated more interest in their future careers, but they also felt that their advisors provided more effective guidance. While our system does rely heavily on the two Advisory College Directors in each college for initiation of student contact, the variety of medical specialties represented by each college's Faculty Affiliate Advisors is able to more satisfactorily address individual career counseling. Although our study was not designed to demonstrate the effect of the new advising system on future career success, we believe that foundations laid in these preclinical years will have a positive impact in the years to come.

Nearly all students in our survey agreed that wellness is important. Students felt that the ACP was more effective at promoting student well-being than the FAP. In general, students felt more comfortable discussing academic stress, physical health, and personal stress with their advisors under the new system. Perhaps, even more important was the finding that students were more comfortable discussing concerns related to their own mental health, thereby allowing for earlier interventions.

We are not alone in our efforts to create an improved advisory system. For example, Murr et al. (2002) describe a five-college system at UCSF (University of California, San Francisco) School of Medicine whose program includes bi-monthly small-group meetings to discuss coping techniques, professionalism, and career options. In addition, one-on-one mentoring is encouraged between advisory directors and their students. Columbia University College of Physicians and Surgeons has created a similar Advisory College System that engages students in bi-monthly meetings and provides additional mentoring as needed (Macaulay et al. 2007). Kalet et al. (2002) describe the implementation of an advisory program at NYU (New York University) School of Medicine whose colleges are theme based and therefore attract faculty members and students with similar interests and goals. One of the fundamental goals of this program is to use the colleges as a structured forum for exercises related to professional development. Finally, Stewart et al. (2007) provide a witty comparison between the four-college advising system at Johns Hopkins University School of Medicine and the four Houses of Hogwarts described in Rowling's (1998) famous Harry Potter series.

Other institutions have developed alternative strategies to enhance student advising and meet other institutional needs. Puckett et al. (1989) describe the establishment of four Advisory Deans at Duke University School of Medicine for the purpose of teaching professionalism and providing career counseling. The University of Washington has created a college system consisting of 30 key Faculty who not only develop one-on-one relationships with their assigned students but are also responsible for teaching clinical skills and professionalism throughout the 4 years (Goldstein et al. 2005). At the University of Wisconsin Medical School, a senior distinguished faculty member is assigned to each medical school class (Scheckler et al. 2004). These mentors attend classes and social functions and are available to students for one-on-one mentoring. Finally, Coates et al. (2008) described a fourth-year medical student College Program based on career interests and designed to help students choose fourth-year electives, promote career mentoring opportunities, and improve the quality of the fourth-year education.

Our study is unique in that it is the first to systematically examine data comparing a traditional one-on-one mentoring system to a structured college advising system that addresses both career counseling and student wellness. Coates et al. (2008) were also able to demonstrate an improvement in students’ abilities to identify a faculty advisor and satisfaction with accessibility of faculty after implementation of a fourth-year college advising system. However, their system focused primarily on career advising and planning, while our study also demonstrated a significant impact on student wellness. Many of the other innovative college advising systems above report that formative surveys and focus groups have shown that students value these interventions (Puckett et al. 1989; Kalet et al. 2002; Scheckler et al. 2004; Goldstein et al. 2005; Macaulay et al. 2007). However, these institutions have not published data regarding overall effectiveness in reaching their goals or compared the results to more traditional advising systems.

We recognize several limitations of our study. First, our study was performed at a single institution, which may limit the ability to generalize our results. Because demographics of our students are comparable to medical school classes across the country, we believe that the institution of a similar college advising system in other schools would be equally successful. We were also unable to achieve 100% response rate, which raises questions as to the concerns or opinions of those students who were unable or chose not to complete the survey. However, we do believe that the striking difference noted between perceptions of advising before and after implementation of the college advising system is not likely to be significantly diminished. Finally, we were limited to measuring students’ perceived effectiveness of the new program, rather than basing our assessment on hard outcomes such as rates of depression or future career success.

We plan to continue to collect annual data regarding student satisfaction with our ACP in an effort to assess whether this improvement in satisfaction persists over time. We also plan to use the students’ responses to further refine the goals of the colleges and the Advisory College Directors. While a majority of students were satisfied with the new system, others did not report as great an improvement in the areas of career counseling as they did in the areas of wellness. We have therefore developed new programs aimed at increasing their exposure to career opportunities and academic advising. We continue to work on strengthening the role of the Faculty Affiliate Advisors within each College system.

While we anticipate that increased student satisfaction with career counseling and increased accessibility to advisors ready to address wellness issues will likely lead to improved student health and overall career success, we need further studies to verify that this is true. We also recognize that there are many innovative mentoring models available, and it will be important to continue to study each new model as it is implemented in order to further refine the best options for undergraduate medical student advising. Ultimately, it is our responsibility to ensure that students embark upon their individual medical careers equipped for success in both their professional and personal lives. We believe that the ACP can function as a significant advisory tool to ensure excellent academic and personal preparation.

Acknowledgments

The authors thank the Advisory College Directors without whom the ACP would not be possible: Oscar Guillamondegui, MD; Heidi Schaefer, MD; Barron Patterson, MD; Elizabeth Sastre, MD; Ban Allos, MD; Amy Fleming, MD; Beau Kelly, MD; and Michael Pilla, MD.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.