Mentors’ perspectives on successes and challenges of mentoring in COG Young Investigator Mentorship Program: A report From the Children's Oncology Group.

Abstract

Background:

Identification and development of Young Investigators (YI) is critical to the long-term success of research organizations. In 2004, the Children’s Oncology Group (COG) created a mentorship program to foster the career development of YIs (faculty <10 years from initial appointment). This study sought to assess mentors’ long-term assessment of this program.

Procedure:

In 2018, 101 past or current mentors in the COG YI mentorship program completed an online survey. Statistical comparisons were made with the Kruskal-Walis Test.

Results:

The response rate was 74.2%. As some mentors had multiple mentees, we report on 138 total mentee/mentor pairs. Mentors were 57.4% male, and mentees were 39.1% male. Mentors rated being mentored as a YI as important with median rating of 90 on 1–100 scale; interquartile range (IQR) 80–100. Most mentors reported that being mentored themselves helped their own success within COG (78.2%) and with their overall career development (92.1%). Most mentors enjoyed serving in the program (72.3%), and the median success rating (on a 1–100 scale) across the mentor/mentee pairings was 75; IQR 39–90. Success ratings did not differ by mentor/mentee gender, but improved with increased frequency of mentor/mentee interactions (p<0.001). Mentor/mentee pairs who set initial goals reported higher success ratings than those who did not (p<0.001). Tangible successes included: current mentee COG committee involvement (45.7%), ongoing mentor/mentee collaboration (53.6%), and co-authored manuscript publication (38.4%).

Conclusion:

These data indicate that mentorship is important for successful professional development. Long-term mentoring success improves when mentors and mentees set goals upfront and meet frequently.

Introduction:

Mentorship is a core feature of research training and involves the dynamic pairing between a junior-investigator (mentee) with a more senior-investigator (mentor) collaboratively working to promote career development. ¹ The mentor often serves several roles, including teacher, sponsor, advisor, agent, role model, coach, and confidante.² In 2004, the Children’s Oncology Group (COG) Young Investigators (YI) committee created the COG YI mentorship program with the aim to increase junior-investigator involvement in COG committees and to promote their career development. The early success, as well as long-term assessment of the COG YI mentorship program by the enrolled mentees has previously been published.^3,4 The current report represents a long-term evaluation of the program from the mentors’ perspective.

Methods:

Overview

The COG consists of more than 200 pediatric institutions in the United States, Canada, Australia, New Zealand, and Europe and conducts biological, therapeutic and supportive care trials across the range of pediatric cancers. In 2004, the COG YI Program was launched to promote career development of junior investigators. The structure of the COG and COG YI committee has been previously described.³ The YI committee is made up of liaisons from 22 of the disease/domain/discipline committees within COG (Table 1).³ Eligible candidates for participation in the COG YI Mentorship Program are junior faculty members (typically within 10 years of completing their terminal degree) who are members of COG in any discipline, including physicians, nurses, pharmacists, psychologists, and other doctoral-level researchers. Applicants complete an application that describes their interests and goals.³ When goals are unclear, a call with a member of the YI mentorship subcommittee occurs to clarify. The application is then forwarded to the liaison from the prospective mentee’s committee of choice, who then contacts and finds the most suitable mentor. Pairings last for three years. During this period, mentees are encouraged to set goals and meet with their established mentor. Over the course of the program, mentees complete an electronic evaluation at three months, one year, two years and three years while mentors complete evaluations at one and three years.

Table 1-.

Children’s Oncology Group Committee Structure¹

Disease	Domain	Discipline	Administrative
1) Acute Myeloid Leukemia 2) Acute Lymphoblastic Leukemia 3) Bone Tumors 4) Central Nervous System Tumors 5) Hodgkin Disease 6) Neuroblastoma 7) Non-Hodgkin Lymphoma 8) Rare Tumors 9) Renal Tumors 10) Soft Tissue Sarcoma	1) Cancer Care Delivery Research 2) Cancer Control and Supportive Care 3) Developmental Therapeutics 4) Epidemiology 5) Outcomes and Survivorship 6) Stem Cell Transplant	1) Adolescent and Young Adult 2) Behavioral Science 3) Clinical Research Associates 4) Cytogenetics 5) Diagnostic Imaging 6) Hematology/Oncology 7) Neuroscience 8) Nursing 9) Pathology 10) Pharmacy 11) Radiation Oncology 12) Statistics 13) Surgery	1) Bioethics 2) Data Safety and Monitoring 3) Diversity and Health Disparities 4) Executive 5) Industry Relations Advisory 6) Institutional Performance Monitoring 7) Membership 8) National Cancer Institute Community Oncology Research Program 9) Patient Advocacy 10) Return of Results 11) Scientific Chairs 12) Voting Body 13) Young Investigators

¹Listed in alphabetical order

Survey

Mentors who participated in the COG YI Mentorship Program for at least one year between March 2004-July 2018 were eligible to evaluate the COG YI mentorship program by completing the Mentor Long-Term Assessment Survey administered via REDcap survey (N=136). As there was not a previously established validated survey that has been used to assess mentorship in a large cooperative trial group, the survey was newly designed and edited by 18 oncology providers who are current members of the COG Young Investigator committee. The 31-question survey included 16 questions specific to mentors’ overall experience with 15 additional questions specific to each mentor/mentee pairing. Completing the survey required approximately 5–10 minutes (Supplementary materials). Non-responders were sent four weekly reminder emails to complete the survey. Mentors were asked about their own experiences as mentees, their experiences as mentors in the program, the success of their mentee pairing, and mentees’ eventual accomplishments. As a program evaluation, the study was deemed IRB exempt. Statistical comparisons were made using the Mann Whitney U and Kruskal-Wallis tests. For continuous variables, median values will be presented followed by the 25^th-75^th interquartile range (IQR) in parenthesis.

Results:

Overall Mentor Cohort

From March 2004-August 2018, 154 unique mentors were identified as serving in the COG YI mentorship program. Twelve had served in the program for <1 year, five were deceased, and one could not be located. The survey was returned by 101/136 eligible mentors for a response rate of 74.2%. Of the respondents, 73 had mentored one YI; 22 had mentored two YIs; and six had mentored three or more YIs in the program. Thus, 138 unique mentor/mentee pairs were captured. In addition, outside of the official YI program, many of the mentors reported being active in mentoring other YIs within COG with 49% (N=50) having 1–6 other mentees, and 14% (N=14) reporting having >7 past mentees. Mentoring that occurred outside of the COG YI Mentorship Program was not assessed in this study. Mentors’ demographics appear in Table 2. At the time of evaluation, most mentors were senior oncologists who trained prior to the year 2000 (66%) and were at the rank of Full Professor (60.4%). The mentors were 57.4% male/42.6% female and were paired to mentees who were 54/138 (39.1%) male and 84/138 (60.9%) female.

Table 2-.

Characteristics of the mentor cohort

Characteristic	n (%)
Sex
Male	43 (42.6)
Female	58 (60.9)
Current academic rank
Assistant Professor	1 (1.0
Associate Professor	33 (32.7)
Full Professor	61 (60.4)
Retired or non-ranked position	6 (5.9)
Year started first academic job
<1980	7 (6.9)
1980’s	22 (21.8)
1990’s	38 37.6)
2000’s	34 (33.7)

Mentor Specific Questions

The 101 mentors were asked about their prior experience as mentees. Seventy-nine (78.2%) mentors reported that they had a mentor who was instrumental to their success within COG. On a scale of 1–100, when asked to rate the importance of this mentorship, the median score was 90 (IQR 80–100). Furthermore, ninety-three (92.1%) mentors said that a mentor had been instrumental to their general career development with a median importance rating of 90 (IQR 80–100). Most respondents enjoyed serving as a mentor in the program (72.3%); 43.6% agreed it had a positive effect on their career and 41.6% felt it helped to hone their own mentorship skills (Table 3). Over 87% agreed that the YI mentorship program should be continued, and 78.2% said that they would be willing to serve as a mentor in the program in the future. The majority of mentors reported a personal direct benefit from the mentoring relationship (60.1%). Representative examples included: having a partner in projects kept mentors excited about (their) work; mentees offered a fresh perspective; mentors benefited from mentee enthusiasm; pairings helped to develop mentorship skills; mentors learned how to be more accessible; pairings led to co-authored publications; and kept the mentor challenged and up-to-date.

Table 3-.

Mentor perceptions of the program

Question	Strongly Agree	Agree	Neutral	Disagree	Strongly Disagree
	n (%)	n (%)	n (%)	n (%)	n (%)
Overall mentor enjoyed serving as mentor in the program	28 (27.7)	45 (44.6)	21 (20.8)	6 (5.9)	1 (1.0)
Overall program had positive effect on your career	15 (14.9)	29 (28.7)	44 (43.6)	10 (9.9)	3 (3.0)
Overall mentorship program has helped to hone your own mentoring skills	10 (14.9)	32 (31.7)	22 (20.0)	15 (14.9)	3 (3.0)
Keeping COG YI mentorship program going	49 (48.5)	39 (38.6)	11 (10.9)	1 (1.0)	1 (1.0)

Mentor and Mentee Pairing Success

When asked to rate the success of the mentor/mentee pairings on a scale of 1–100, mentors for all 138 mentor-mentee pairings rated it at a median of 75 (IQR 39–90). The success of these pairings varied according to how frequently the mentor and mentee interacted by phone or in-person for the duration of the pairing (p<0.001) (Table 4). Setting goals at the start of a pairing occurred in 57.2% of the pairings and goals were achieved in 49.2%. Setting goals at the start of the mentor-mentee relationship significantly increased the success rating of the pairs (p<0.001) (Table 4). In 37% of the pairings, no major barriers to the mentoring process were reported. Identified barriers in other pairings included: mentors lacked a project in the YIs chosen field (23%), mentor was too busy (18%), mentee was too busy (15%), mentee was not engaged/did not follow through (10%), mentee had unrealistic expectations (8%), mentee lacked the necessary research skill-set (5%), personality differences resulting in a poor mentor-mentee fit (4%), mentee was unable to attend COG meetings (4%), mentee was a poor medical writer (3%), and other (22%) e.g., geographic distance between mentor/mentee, mentee changing careers, competing need to mentor junior investigators at mentor’s own institution, and lack of funding. Success of the pairings were not statistically significantly different according to the gender within the mentor/mentee pairing (male mentor/male mentee [n=38] median success 75 [IQR 48–90], male mentor/female mentee [n=40] median success 75 [IQR 50–90], female mentor/male mentee [n=16] median success 80 [IQR 35–89], and female mentor/female mentee [n=44] median success 73 [IQR 30–85]).

Table 4-.

Mentor/mentee pairing success according to frequency of interaction and setting initial goals

Question
	N (%)	Median success rate1 (IQR)2	P Value
Frequency of mentor/mentee interaction			P<0.0013
Never/only once	17 (12.3)	0 (0–11)
Yearly	12 (8.7)	50 (34–58)
Twice yearly	29 (21.0)	69 (28–75)
Three-four times a year	47 (34.1)	80 (70–90)
Monthly	27 (19.6)	95 (80–100)
Weekly	6 (4.3)	95 (85–100)
Goals were set at the beginning of the mentor/mentee pairing			P<0.0013
Strongly Agree	23 (16.7)	95 (81–100)
Agree	56 (40.6)	80 (50–90)
Neutral	27 (19.6)	72 (40–83)
Disagree	24 (17.4)	50 (20–74)
Strongly Disagree	8 (5.9)	0 (0–1)

¹On a scale of 1–100.

²Interquartile Range

³Kruskal-Wallis tests

Specific success outcomes

Mentors reported assisting their mentees in several ways during the official pairing period, including providing guidance on specific research projects (50.7%), offering general career advice (47.1%), giving specific disease committee advice such as introduction to committee leadership (39.1%), collaborating on manuscripts for publication (21.7%), reviewing/assisting with a grant application (9.4%), assistance engaging with a non-COG consortium (8.7%), employment assistance (6.5%); only 10.1% of mentors reported not being able to help their mentee. Mentors also reported tangible successes for the mentees that resulted from the pairings, including becoming active in a COG [steering or study] committee (45.7%); ongoing collaboration with the mentee (53.6%); established new research collaboration (40.6%); national/international presentation (39.9%); manuscript publication (38.4%); grant application (18.1%); and funded grants (15.2%).

Discussion

These data demonstrate that many of the over 100 mentors who have served in the COG YI mentorship program feel that their specific pairing was successful (median success score 75) and has led to tangible success for many of the mentees. This indicates that mentoring through a cooperative group is feasible and, in many cases, leads to long-term success. These data are consistent with the previously published COG Long-term Assessment from the past COG YI mentees’ perspective that showed a median success rating of 72.³ Other groups such as American Society of Pediatric Hematology Oncology (ASPHO) and the American Society of Hematology (ASH) have also demonstrated the success and significant challenges of mentorship within larger societies in hematology-oncology.^5,6 The COG program is unique in that it shows mentoring someone from another institution is possible and the measurable outcomes can be sustained well beyond the initial three-year pairing.

These data also confirm that mentorship is a critical component to the success of the careers of mentors themselves when they were junior investigators. Furthermore, over 60% of mentors felt that they derived a direct benefit from participating and mentoring in the COG YI mentorship program. In addition to personal satisfaction, mentors participating in the COG YI mentorship program are officially recognized for their service and eligible to receive Excellence in Mentoring award. Such recognition for mentoring is often lacking in academic practice.⁷

As expected, meeting frequently was significantly associated with greater success in mentor/mentee pairings. Previously, our program showed that meeting at least quarterly increased academic productivity; however, the current analysis indicates that more frequent interaction may be beneficial.⁴ This finding confirms the results of other studies that suggest that the best mentors are easily accessible and have regular, frequent, and high-quality mentor-mentee meetings.^8,9 Setting goals and expectations at the start of the mentorship relationship was also shown to be associated with more highly rated mentorship pairings. This finding is supported by mentorship literature that has shown mutual goal setting to be a key component of successful mentoring.^10–12 Among low-rated pairings, lack of commitment by the mentee or the mentor was a common theme. This is consistent with Cho et al. who showed successful mentoring was characterized by reciprocity, mutual respect, clear expectations, personal connection and shared values, while unsuccessful mentoring was characterized by poor communication, lack of commitment, personality differences, perceived competition, conflicts of interest, and mentor’s lack of experience.⁸ Compared to some other mentorship studies that have shown low female participation, female mentors and mentees were well represented in this cohort (42.6% and 60.9% respectively).¹¹ There were no significant differences in the success of the pairings according to gender. This is in contrast to many past mentorship studies that have shown female mentees being associated with difficulty finding mentors and achieving success.¹¹

Currently, the COG YI mentorship program has 61 active pairings and has already used the survey data included in this report to improve the mentorship program by adding a list of expectations to the initial mentorship packet (Supplementary materials), including a specific list of goals in the electronic evaluations, and designing mentorship educational sessions for new/active mentors in the program.

Conclusion:

Mentors who have served in the COG YI program reported mentorship has been important to their own success in COG and have shown a strong commitment to mentoring future COG YIs. Long-term mentoring success improves when mentors and mentees set goals and meet frequently. Moving forward, efforts to promote this strategy by future mentor mentee pairings will be pursued.

Abbreviations key

ASH

American Society of Hematology

ASPHO

American Society of Pediatric Hematology Oncology

COG

Children’s Oncology Group

IQR

Interquartile Range

YI

Young Investigators