The double whammy: Advanced medical training and parenting

Abstract

Clinicians may become parents during their clinical training and may be exposed to several challenges in career development, burnout and work-life balance. Previous research findings have reported that stressors facing trainees with children warrant greater attention from graduate medical institutions. Additionally, parenting-related information and considerations about the needs of trainees with children across clinical specialties are needed to inform institutional and national policies. A quantitative approach was used to examine clinical trainees’ perceptions and experiences of parenting in relation to different specialties, sociodemographic traits, levels of support, and other potential factors influencing their residency and fellowship training and well-being. We used a survey that was distributed to all University of Toronto medical trainees (2214) via email correspondence and social media platforms. The trainees were asked to base their answers on their experience during the academic year of 2019 to 2020 (before the Coronavirus Disease 2019-related shutdown). Our study revealed that clearly, burnout is a concern for physicians who are raising children while in training. Notably, it was higher among younger aged trainees and those beginning their training journey including, first-year fellows and second-year residents, in addition to parents with toddlers. Moreover, female residents and male fellows showed higher burnout than their counterparts. Institutional support was associated with lower rates of burnout, evidenced by access to opportunities, allowing time to breastfeed/express milk and having access to medical care. We found multiple independent and significant factors affecting their rate of burnout including limited access to opportunities, lack of a self-care routine and absence of social community outside of work.

The results show the importance of creating a positive learning experience for trainees juggling parenting and training needs, especially those starting their training both as residents and as fellows and those with younger children. Interventions can be categorized into those targeted at the individual and family levels, and institutional levels, with the overarching goal of balancing training and parenting. This can be achieved by fostering learning environments that prevent and decrease burnout and enhance the well-being of trainees and their families, which can start with ensuring trainees are aware of available resources and possible accommodations.

1. Introduction

Clinicians may become parents during their clinical training and may be exposed to several challenges including career development, burnout and work-life balance issues.^[1,2] Experts and previous research findings have reported that stressors facing trainees with children warrant greater attention from graduate medical institutions.^[3,4] In addition to being physicians, residents often have to fulfill other life aims, such as parenting, especially since residency usually coincides with their twenties and thirties, which are the crucial reproductive years.^[5] This can lead to conflicting burdens, which could potentially add to the stress of residency.^[6]

Trainee perceptions and views about parenting challenges and resources are many and not completely understood.^[7] Additionally, parenting-related considerations and information about the needs of trainees with children across clinical specialties are needed to inform institutional and national policies. Data in the area of parenting during residency is limited and controversial, with some data showing the protective nature of having children during residency and other data showing the potential of burnout risk due to competing demands. Residency programs often have clear policies for parental leaves, but most do not have support systems in place beyond the demands of pregnancy, parental leave and breastfeeding.^[8]

Hence, this study aims to explore the perceptions and experiences of parent-clinical-trainees across different medical and surgical specialties, sociodemographic traits, levels of support, and other potential factors influencing their residency and fellowship training and well-being using a quantitative approach. Our literature review of multiple databases led us to develop a level of insightful framework guiding our methodological approach. We found that exploring the phenomenon of parenting and its effect on clinical training is multi-dimensional, hybrid and categorical nature. These conceptual elements have spurred a comprehensive approach to explore such dimensions and we intended to be inclusive of all possible influential variables.

2. Methods

In our study, we used a quantitative cross-sectional approach. Approval of the Research Ethics Board at the University of Toronto was obtained. The sample size (m) was determined by the target precision of the estimate of burnout using the following equation: m = anticipated prevalence × (1-anticipated prevalence)/SE². With a target precision of 10%, if the anticipated rate of burnout is assumed to be around 30% (i.e., 0.3 in the above equation), based on previous studies conducted,^[9–11] and with an aim of 95% confidence interval (CI) of 25% to 35% (anticipated prevalence ± 1.96SE), 200 completed questionnaires would allow us to approximate the true estimate. Assuming a response rate of 50% and a completion rate of 80%, doubling the sample to 450 was decided.

The survey included an online questionnaire that was developed by the authors after a literature review and several discussions between authors to finalize all the possible questions that should be included in the questionnaire appropriately for a diverse multinational population. It included several items: part 1 involved the sociodemographic (e.g., age, gender, number of children), Part 2 consisted of 6 Likert-type questions regarding stress and job satisfaction in training using Likert scales using a 4-point categorical scale from strongly agree to strongly disagree (e.g., childcare, income, parenting experiences while studying, supports needed and barriers when support is needed, their management of their well-being and health in the dual role of parenting and studying as post-graduate medical trainees). The third part included the Maslach burnout inventory scale measured on a 7-point Likert scale ranging from 0 to 6, with scores summed and categorized into published instrument standards of “low,” “moderate,” and “high” scores in each subscale category.^[12] However, we used only the 2 items combined (Burnout and callous toward people) selecting always/daily or/and almost always, for diagnosing the rate of burnout as previously recognized by several researchers.^[13,14] Nevertheless, we reported all other items of the burnout Maslach scale. The trainees were asked to base their answers on their experience during the academic year of 2019 to 2020 (before the Coronavirus Disease 2019-related shutdown).

The survey was piloted on 20 trainees and minor changes were made to the wording of some questions, and their feedback helped in improving the clarity of the questions. We intended to include all post-graduate clinical trainees as our sample population and hence, all registered residents and fellows were emailed with the online survey, for completion. The survey took an average of 10 minutes to complete. The participants were contacted via multiple modalities: email of Post Graduate Medical Education distribution list of all currently registered residents and clinical fellows (2214) at the University of Toronto, advertisement on social media (Post Graduate Medical Education Twitter account), WhatsApp groups and Facebook of parents’ groups (Canadian physicians’ moms, Greater Toronto Area Medical Doctors moms, Physician parents for gentle and respectful parenting, Canadian Physician health staff and Canadian women psychiatrist group). The trainees were self-identified by responding to the email or advertisement. Inclusion criteria were: any resident or clinical fellow at the University of Toronto from all medical departments, with 1 or more children. Those who were currently pregnant, with no children at home or their children, were living outside of the participant home ≥ 90% of the time were excluded from the study. Statistical analysis was done using the statistical package for social sciences IBM SPSS version 22.^[15] Categorical variables were presented as frequencies and percentages and were compared using the chi-squared test or Fisher exact test depending on the context. The Mann–Whitney U test was used for parental leave and Likert-scale comparison, and Pearson chi-square test was used for the remaining comparisons. P values <.05 were considered statistically significant. A multivariable logistic regression analysis was performed to determine if there are any associated factors for the development of burnout among the total sample or among the separate groups of fellows and residents. The results of regression analysis were presented as odds ratio (OR) with 95% CI. The significance level was set at P value ≤ .05 for all statistical tests. The online data was protected by online security measures provided by Qualtrics XM.^[16]

Participants’ information was kept within password-protected computers on a secure server in the Faculty of Medicine at the University of Toronto and only accessible to the study team. Consent was obtained at the beginning of the survey with an information sheet explaining the purpose and benefits of taking part in this study. We also assured participants that the information they share will not be connected to them directly in any way and their participation in the study will not be disclosed to their program or their colleagues. Furthermore, they were not identified as per their specific programs, as they were identified as medical or surgical trainees. This was done to avoid any risk of pointing out those with children, especially in small programs. There were no limitations to withdraw from any component of this study at any time without any negative consequences or any change to their relationship with the University of Toronto. There is potential discomfort in disclosing residency experiences. Another theoretical risk was anticipated if someone could be recognized by describing a certain situation, but we made all efforts available to remove all classified information from any participants so that no one could be inadvertently identified. Additionally, to fulfill policy requirements, researchers must de-link identifying information, encrypt personally identifiable information and/or allow personnel access to data on a need-to-know basis. No funding is needed as participants will be asked to volunteer their time to participate in the study.

3. Results

There were 450 participants, and they represented those who fulfilled our study inclusion criteria. Respondent demographic information is presented in Table 1. 51.7% (219) of our participants were female and 48.3% (205) were male. The majority (55%, 235) were young (26–35 years old), 41% (178) were aged 36 to 45 years and the rest (3.35%, 14) were aged ≥ 46 years. 401 (94.6%) were married, 16 (3.8%) had partners in common law (3.8%), 2 (0.5%) were divorced, 2 (0.5%) were single and 3 participants did not provide their marital status. 46.65% (199) of participants had 1 child living with them, 35.4% (151) had 2 children, 11% (47) had 3 children and 3.3% (14) had 4 children and 1.25 (5) had 5 children. The rest 2.6% (11) did not have children living with them. 51.3% reported that their children have health concerns. They differed in their level of training, specialty and duration of their living status in Toronto (Table 2). The majority of participants (62.2%) were enrolled in medical training programs and 23.1% were enrolled in surgical training programs.

Table 1.

Participants’ sociodemographics.

		N (427)	%	P value
Gender	Female	219	51.7	.528
	Male	205	48.3
Age	26 to 35	235	55.0	<.0001
	36 to 45	178	41.7
	46 or older	14	3.3
Marital status	Single	2	0.5	<.0001
	Common law	16	3.8
	Married	401	94.6
	Divorced	2	0.5
	Other	3	0.6
How many children live with you?	Do not live with me	11	2.6	<.0001
	1	199	46.6
	2	151	35.4
	3	47	11.0
	4	14	3.3
	5	5	1.2
Infant	0	309	72.4	<.0001
	1	117	27.4
	2	1	0.2
Toddler	0	247	57.8	<.0001
	1	172	40.3
	2	8	1.9
Early child	0	330	77.3	<.0001
	1	74	17.3
	2	23	5.4
Mid child	0	302	70.7	<.0001
	1	82	19.2
	2	35	8.2
	3	8	1.9
Adolescence	0	400	93.7	<.0001
	1	19	4.4
	2	7	1.6
	3	1	0.2
18 and more	0	417	97.7	<.0001
	1	7	1.6
	2	3	0.7
When you had your first child?	Prior to residency/fellowship	260	60.9	<.0001
	During residency//fellowship	167	39.1
When you had your last child?	Prior to training	24	30.4	.001
	During training	55	69.6
Including yourself, how many income earners do you consider contributing to household?	1 adult	143	35.8%
	2 adults	248	62.2%	<.0001
	More than 2 adults	8	2.00%
What was your total 2018 household income?	Less than $55,000	32	8.0%	<.0001
	55,000 to < 85,000	117	29.35
	$ 85,000 or more	213	53.4%
	Prefer not to say	37	9.3%

Table 2.

Participant’s description of training status.

Measures		N	%	P value
How many yr have you lived in Toronto or the GTA in total?	<1 yr	110	26.3%	<.0001
	1–2 yr	85	20.3%
	3–4 yr	48	11.5%
	5–6 yr	46	11.0%
	7–10 yr	46	11.0%
	More than 10 yr, but not my whole life	63	15.1%
	My whole life	20	4.8%
Level of training?	PGY1	20	4.7%	<.0001
	PGY2	36	8.5%
	PGY3	37	8.7%
	PGY4	38	9.0%
	PGY5	54	12.8%
	PGY6	8	1.9%
	PGY7	3	0.7%
	PGY8 or higher	2	0.5%
	Clinical fellow—yr 1	108	25.5%
	Clinical fellow—yr 2	64	15.1%
	Clinical fellow—yr 3	25	5.9%
	More than 3 yr	28	6.6%
Specialty	Medicine	258	62.2%	<.0001
	Surgery	96	23.1%
	Other	61	14.7%

GTA = Greater Toronto Area.

The rate of Burnout items varied in the Likert scales of 6-point categorical measures that ranged from always/daily to never (Table 3). The overall prevalence of burnout was 31.7% among female trainees and 29.5% among male trainees. Burnout was higher among younger-aged trainees (38.2%), married (59.9%), parents with more toddlers, those enrolled in medical specialties, first-year fellows (11.2%) and residents in their second year of training (9%) (Table 4).

Table 3.

Rate of burnout among participants.

	Always/daily	Almost always	Sometimes	Rarely	Almost never	Never
Physical exhaustion	65	151	171	30	7	2
	15.3%	35.4%	40.1%	7.0%	1.6%	.5%
Emotional exhaustion	44	97	212	49	16	8
	10.3%	22.8%	49.8%	11.5%	3.8%	1.9%
Overwhelmed	44	96	210	53	18	5
	10.3%	22.5%	49.3%	12.4%	4.2%	1.2%
Burned out	40	83	202	54	31	16
	9.4%	19.5%	47.4%	12.7%	7.3%	3.8%
Callous toward people	7	17	169	117	60	56
	1.6%	4.0%	39.7%	27.5%	14.1%	13.1%
Lonely	10	49	181	100	43	43
	2.3%	11.5%	42.5%	23.5%	10.1%	10.1%
Extremely anxious	20	54	201	92	45	14
	4.7%	12.7%	47.2%	21.6%	10.6%	3.3%
Happy	32	157	212	21	3	1
	7.5%	36.9%	49.8%	4.9%	.7%	.2%

Table 4.

Prevalence of burnout.

		Fellows	Residents	Total
Gender	Female	11.2%	20.5%	31.7%
	Male	17.5%	12%	29.5%
Age	26–35	13.7%	24.5%	38.2%
	36–45	13.7%	8%	21.7%
	46 or older	0.9%	0%	0.9%
Marital status	Single	0%	0.5%	1%
	Common law	1.3%	1%	2.3%
	Married	27.4%	29.5%	56.9%
	Divorced	0%	1%	1%
	Other	0%	0.5%	1%
How many children live with you?	3 and less	22.1%	31%	53.1%
	More than 3	6.2%	1.5%	7.7%
Specialty	Medical	16.5%	23%	39.5%
	Surgical	6%	9.7%	16%
	Other	6.4%	0%	6.4%
Level of training?	PGY1		2%	2%
	PGY2		9.6%	9.6%
	PGY3		5.6%	5.6%
	PGY4		6.6%	6.6%
	PGY5		6.6%	6.6%
	PGY6		1.5%	1.5%
	PGY7		0%	0%
	PGY8 or higher		0.5%	0.5%
	Clinical fellow—yr 1	11.2%		11.2%
	Clinical fellow—yr 2	8.9%		8.9%
	Clinical fellow—yr 3	4.9%		4.9%
	More than 3 yr	3.6%		3.6%

PGY = Post Graduate year.

Parents had childcare support during the daytime from multiple sources (Fig. 1): partners (43.8%), daycare provider (39.1%), school (24.1%), family member (17.3%), live out-child care provider (14.5%), live-in child care provider (4.2%) and few did not require any help with childcare. However, during their evening working hours, their children support was provided by their partners (80.1%), family members (23.7%), paid-by-hour sitters (9.1%) or in-house care sitters (6.5%).

Figure 1.

Percentage of parenting factors affecting participants’ training.

Survey results revealed that participants strongly agreed (53.6%) and somewhat agreed (37.75%) that they were generally healthy (Table 5). However, few strongly/somewhat disagree (7.7%/0.9%) with being generally healthy. Also, some of the residents/fellows believed strongly (2.8%) or somewhat (6.6%) that their health conditions negatively impacted their training. Additionally, many somewhat strongly (45.2%) and strongly (12.6%) agreed that training experience is negatively affected by parenting status. Participants’ lifestyles and social activities varied in their degree of commitment. However, many were active and had some sort of self-care routine (Table 5). Additionally, they addressed that they had social, healthcare access and institutional support when needed. However, the majority (61.8%) were not aware of the resources provided to parenting residents at the University of Toronto. Many were stressed due to the need to strictly limit their work hours in order to allow them to pick up/drop off their child/children to/from childcare; as well as financial concerns that restricted their spending on childcare services.

Table 5.

Percentage of the Biopsychosocial factors influencing parents’ training.

	Strongly agree	Somewhat agree	Somewhat disagree	Strongly disagree
Generally healthy.	229	161	33	4
	53.6%	37.7%	7.7%	.9%
Health condition that negatively impacts training	12	28	68	319
	2.8%	6.6%	15.9%	74.7%
Training experience is negatively affected by parenting status.	54	193	113	67
	12.6%	45.2%	26.5%	15.7%
Can exercise at least weekly.	63	102	105	157
	14.8%	23.9%	24.6%	36.8%
Self-care routine.	40	182	135	70
	9.4%	42.6%	31.6%	16.4%
Have a social community outside of work.	105	164	115	43
	24.6%	38.4%	26.9%	10.1%
Person can turn to in emergency.	171	167	61	28
	40.0%	39.1%	14.3%	6.6%
Person can count on.	202	158	42	25
	47.3%	37.0%	9.8%	5.9%
Emotional support.	117	191	89	30
	27.4%	44.7%	20.8%	7.0%
Opportunity to spend time with family.	127	196	67	37
	29.7%	45.9%	15.7%	8.7%
Getting the support, I need from my training program.	88	175	107	57
	20.6%	41.0%	25.1%	13.3%
Have access to medical accommodations at work if I need them.	77	173	104	73
	18.0%	40.5%	24.4%	17.1%
Aware of the resources provided to parenting residents at University of Toronto.	7	40	116	264
	1.6%	9.4%	27.2%	61.8%
Stressed for limiting work h due to childcare drop off and/or pick up.	140	172	64	51
	32.8%	40.3%	15.0%	11.9%
Household income is sufficient.	51	121	133	122
	11.9%	28.3%	31.1%	28.6%

Moreover, they commonly had acceptable time to commute to work; 50.6% needed ≤ 30 minutes and 36.8% needed between 30 and 60 minutes (Table 6). We also found a lower rate of burnout in residents with fewer toddlers (P = .03, CI = 0.023–0.036) or fewer children with health concerns (P = .005, CI = 0.002–0.012).

Table 6.

Percentage of other factors influencing parents training.

Factors		N	%	P value
How long is your current commute to work?	<30 min	202	50.6%	<.0001
	30–< 60 min	147	36.8%
	1–2 h	41	10.3%
	More than 2 h	9	2.3%
Limited access to opportunities	No	308	72.1%	<.0001
	Yes	119	27.9%
Being overlooked in promotions or leadership positions	No	393	92.0%	<.0001
	Yes	34	8.0%
*Limited access to breastfeeding rooms	No	124	72.7%	<.152
	Yes	18	27.3%
*Not being given time to breastfeed	No	136	81.8%	<.128
	Yes	12	18.2%

^*Last 2 questions were analyzed for women only.

Results showed supportive factors were significantly associated with a low rate of burnout, and these included: a perceived sense of good general well-being (among residents; P = .001, CI = 0.0003–0.005, fellows; P = .008, CI = 0.003–0.012) with less health conditions affecting their training (residents only: P = .024, CI = 0.019–0.028). Other individual and social protective factors against burnout were: having a self-care routine (Residents; P = .005, CI = 0.0008–0.009, fellows; P = .015, CI = 0.008–0.024), availability of social community outside work (Residents; P = .020, CI = 0.015–0.024, fellows; P ≤ .001, CI = 0.001–0.00001), presence of emotional support (Residents; P = .021, CI = 0.006–0.035, fellows; P ≤ .001, CI = 0.001–0.00001), opportunity to spend time with the family (Residents; P = .002, CI = 0.0003–0.006, fellows; P ≤ .001, CI = 0.001–0.00001) and having a sufficient income (Residents; P = .024, CI = 0.019–0.028, fellows; P = .008, CI = 0.003–0.012) (Table 7). However, only residents had higher rate of burnout related to the high levels of stress that accompanied the task of drop-of and/or pick-up from childcare (P = .018, CI = 0.013–0.022).

Table 7.

A multivariate analysis of the independent and significant associations between burnout and all possible variables.

Variables	B	P value	OR	95% CI for OR
				Lower	Upper
Limited access to opportunity	0.748	.004	1.473	1.285	1.786
Generally healthy	−0.855	.033	0.425	.194	0.932
Training experience is negatively affected by parenting status	0.933	001	2.542	1.497	4.316
Self-care routine	−0.574	.018	0.563	.350	0.906
Have a social community outside of work	−0.509	.042	0.601	.368	0.982
Emotional support	−0.134	.622	0.875	.514	1.490
Opportunity to spend time with family	−0.500	.062	0.607	.358	1.026
Getting the support I need from my training program	−0.298	.238	0.742	452	1.218
Household income is sufficient	−0.336	.200	0.715	.428	1.194
Constant	1.161	.026	3.192

CI = confidence interval, OR = odds ratio.

Furthermore, in terms of parents’ training support, among residents and fellows, 72.1% acknowledged that they had access to opportunities and 92% denied being overlooked in promotions or leadership positions due to their parenting status. The majority expressed that there was no limited access to breastfeeding rooms (72.7%) and were given enough time to breastfeed/express milk (81.8%) (Table 6).

Bivariate analysis revealed some significant associations between burnout and several possible correlation factors among participants. Results revealed that training experience is negatively affected by parenting status in both residents and fellows and was significantly associated with a higher level of burnout (P ≤ .001, CI = 0.001–0.00001).

Institutional support was associated with fewer rates of burnout, evident by access to opportunities (residents; P = .01, CI = 0.03–0.052, fellows; p=<0.0001, CI = 0.001–0.00001), allowing time to breastfeed/express milk among residents only (P = .036, CI = 0.010–0.051), getting support from the training program (Residents; P = .002, CI = 0.0003–0.006, fellows; P = .001, CI = 0.0003–0.005) and having access to medical care (Residents; P = .014, CI = 0.009–0.018, fellows; marginally significant P = .051, CI = 0.046–0.0551).

The backward multi-step regression analysis in the total sample of residents and fellows revealed multiple independent and significant factors affecting their rate of burnout. For example, limited access to opportunities (P = .004, OR = 1.473 [1.285–1.786]) doubles their rate of burnout. Moreover, their belief of the fact that training experiences are negatively affected by parenting status also significantly and dependently increased their burnout (P = .001, OR = 2.542 [1.497–4.314]). Lack of a self-care routine was also found to be of importance and negatively associated with a higher rate of burnout (P = .018, OR = 0.563 [0.350–0.906]), as well as the absence of social community outside of work (P = .042, OR = 0.601 [0.368–0.982]) (Table 7).

4. Discussion

Our study revealed interesting findings exploring the rate of burnout and possible sociodemographic, biopsychosocial, individual and institutional factors. Whilst direct comparison of burnout scores in our study to previous studies may be challenged due to the use of different scales, it is interesting that similar factors have been found to be associated with higher levels of burnout. Factors that increase trainee risk of burnout differ by study, with limited studies providing reproducible results.^[17,18]

Our bivariant regression results revealed significant associations between burnout and sociodemographic profiles for the residents with toddlers and trainees with children with health concerns, which might be explained by the possible increased attention needed for these groups of children. Limited significant associations were found for other sociodemographic profiles and 1 previous study showed similar results of a lack of such correlations.^[19] Similarly, in a systematic review, researchers found that no individual factors, such as age, sex, or the number of years in training, were related to burnout.^[20] Nevertheless, other studies reported that age (younger trainees) and gender (females) are related to a higher rate of burnout.^[21–23] This can be explained by the different methodologies and scales assessing the level of burnout in different settings and countries and the bias of self-reporting.

Our study revealed that burnout is higher among younger age groups (26–35 years old). This is in accordance with previous studies.^[24,25] This could be explained by fewer years of experience, coping skills, financial status and living situation. However, older age does not certainly go hand in hand with extended professional experience and it is just as plausible that greater life experience is responsible for this correlation.

It is well predicted that burnout is highest during the residency period and subsequently declines as the graduate attains more senior positions.^[26] This is likely a result of the high assignment and constant learning strains on the trainees, which ease to some extent as they become experienced trainees. Conferring to the well-known demands-control model by Karasek,^[27] job control is anticipated to moderate the relationship between job demand and psychological strain, which could be a probable explanation for these observations, assuming that the higher biological age goes with more work experience, which in turn leads to more control over the work atmosphere. Our results showed that parents in their earlier training years had the highest prevalence of burnout. Early training years were found to be associated with a high rate of burnout.^[28] Symptoms of burnout often begin in medical school and progressively worsen throughout training, increasing gradually in the first year of residency training and deteriorating into the second year.^[29] A previous study found that the prevalence of burnout ranged from 22.2% in new Post Graduate year4 residents to 83.3% in residents at the end of Post Graduate year2.^[30]

Our results also revealed that female residents reported a higher rate of burnout than their male equivalents. This too is similar to previous studies showing that females had higher levels of burnout (30% vs 15%, P = .014) and emotional exhaustion (22% vs 9%, P = .005)^[9,21,31,32] which could be attributed to conflicting personal and professional demands as well as sociocultural expectations for women. When looking at the relation between marital status in our study population and level of burnout, it initially appeared to us that the married trainees had a higher level of burnout, however, this soon was discovered to be an inconclusive finding since 94.6% of our sample were married. Furthermore, the relational aspects were examined by Warde et al^[33] who aimed to measure “personal and professional” factors that influence parental and marital satisfaction in physicians, and it was found that the level of satisfaction is higher with more supportive partners and less role conflict for both men and women. A more recent study by Adám et al,^[34] aimed to identify the relation of social supports with work-family conflict, in female and male physicians and showed a gender difference, as female physicians were found to have a higher level of work-family conflict compared to males, with the social supports, including parental, peer and organizational supports, believed to be a major contributor to this conflict. Another study focusing on gender differences, based in Serbia,^[35] revealed that female physicians do more childcare and housework than their male counterparts. Having said that, there was no significant difference identified between genders, for Work to Home interference and Home to Work interference, but the Work to Home interference was higher in both genders than the Home to Work interference. In this study, it was identified that cultural background plays a major role in the perception of gender roles, and the effect that has on physicians and their different roles. In a multinational large study, it was found that work–life balance frustration was associated with increased burnout (OR 4.5, P < .001).^[10] Perceived limited training opportunities and feelings of being unappreciated are similar to the previously reported article.^[36]

Furthermore, work-related factors have a greater effect on trainee burnout and psychological well-being than non-work-related factors.^[37] In our results, there was a higher rate of burnout among medical than surgical trainees. This could be explained by the fact that our sample of medical trainees (258) was higher than the surgical trainees (96), resulting in the ambiguity of its prevalence. This is in contrast to previous studies where they found that surgical residents had a high personal burnout.^[38,39] Researchers hypothesized that surgical residents were exposed to greater stress during their training, leading to an increase in the rates of burnout.^[40,41]

In our study, residents with toddlers had a higher level of burnout than fellows with toddlers or trainees with older children. When reflecting on possible causes, it is possible that residents are more likely to be new parents during their training, and learning to juggle both parenting and trainee roles for the first time, and managing the different work demands including oncalls and training program requirements. In addition, being a parent to a toddler requires a lot of patience and learning new skills, which the new parent would not have required before having the child. Remarkably, in a study based in Serbia, it was found that physicians (both men and women) with children aged > 12 years old had higher Work to Home interference than physicians with younger children; however, it is important to note, that this sample of physicians were not expected to do on-call duties.^[35] Furthermore, parents with children with health concerns had a higher rate of burnout, which is expected given the high demand required by trainees in these cases.

When exploring social and institutional supports and their effect on our sample, we found that training experience is negatively affected by parenting status in both residents and fellows and was significantly associated with a higher level of burnout and institutional support was associated with fewer rates of burnout, evident by access to opportunities, awareness of resources and allowing time to breastfeed/express milk. In a study from The Canadian Association of Postdoctoral Scholars, the average Canadian postdoctoral trainee age was 34 years.^[42,43] 48% of the trainees were women, and one-third (31%) had dependent children.^[43] Contrary to other studies, this sample did not elicit major gender differences in the need to have a balanced work-personal life. However, it is apparent here that “parenting-friendly” residency programs would provide support for those residents through this demanding time of life, but additional information would need to be collected as to what “a parenting-friendly residency” is. Trainees who had positive coping approaches available were shown to have a lower prevalence of burnout; practices such as communicating with family, social interaction with friends, or increasing levels of exercise and self-care were reported to be effective coping mechanisms to decrease the effect of burnout.^[44,45] Similar to previous studies, we found that a lack of community and institutional support is linked to an increase in the rate of burnout.^[46,47] Institutional support could have a positive impact on the rate of burnout and trainees’ well-being.^[48,49] Access to mental health services (OR, 0.14; 95% CI, 0.04–0.47) was associated with lesser odds of burnout.^[50] These researchers also linked financial concerns and income level to higher odds of psychological and emotional symptoms.

In a study by Morris and colleagues, in one of the Midwestern Universities in the United States, they explored parenting minor children and found that the perception of basics affecting the residency experience was variable, with positive elements being “family-friendly social activities and support for breastfeeding and negative elements being occupied schedules, “flipflop days and nights and “doubt around how to access sick leave.”^[51] In addition, residents expressed both positive and negative outcomes of being parents; positive outcomes being the pleasure of parenthood and learning how to “draw a line” between work and home, and negative outcomes were being remorseful feelings when having to take time off and work-related tasks being pushed to a later time if more time was chosen to spend with family.

Generally, physicians in residency are at higher risk of depression, poor physical health, anxiety, and suicide.^[52] This can be due to increased demands of residency training, which can often have detrimental effects on the overall health status, thus reducing the resident physician general wellness. Residents experiencing burnout or decreased wellness may also have declined empathy, may have lower adherence to healthcare guidelines or may be more likely to make errors in the provision of patient care. Regularly experiencing high levels of stress can lead to challenges to trainees’ mental and emotional wellness and affect the exacerbation of primary mental illnesses.^[53] Poor physician trainees’ wellness and high rates of burnout affect the healthcare system on many levels. Those experiencing burnout frequently experience decreased physical health, mental health, and general job satisfaction.^[54] In our study, high proportions of our participants expressed always/daily and or almost always being “physically exhausted,” as well as “unhappy and lonely” sometimes. We found that when it comes to health and wellness, supportive factors were significantly associated with a low rate of burnout, and these included a perceived sense of good general well-being, fewer health conditions affecting their training and having a self-care routine. Furthermore, according to the majority of our participants, having a social network outside of work, as well as spending time with loved ones and reliable friends, are protective against burnout.

4.1. Study strengths

Our study has several strengths including the large heterogeneous sample we recruited, and the various specialties of the trainees (both residents and fellows). Moreover, the enormous multi-faceted factors (individual, bio-psycho-social and institutional) that could be associated with burnout were investigated. We also used a well-validated scale to measure burnout. We are hoping that our contribution will add to the working knowledge and support that can be developed for residents and fellows with children and their well-being. We thought that our study might generate theory-based approaches to the implications for trainees’ support and tolerance to life stressors while training and parenting. Our results helped expose clinical trainees to an appreciation of the influences of gender, income, level of clinical experiences and multiple social and institutional policies on individual trainees and their families. We believe this ecological understanding is essential to effective assessment and supportive intervention. Also, learning about the interaction between trainees’ professional and social responsibilities, can alter their own paradigms and bring a supportive understanding to their struggles during clinical training and parenting. Further positive environmental support and reduced burnout rates will result in fewer trainee absences, a lower attrition rate of them leaving the program, alongside progression in both economic and patient safety-related outcomes.^[55] Hettler and colleagues proposed the Six Dimensions of Wellness Model, which is a framework highlighting the interconnected facets of wellness affecting the individual overall wellness status.^[56] Such effective interventions will support several dimensions of wellness through providing resources or opportunities for trainees, helping them develop tools and strategies to support their own wellness, or eliminating organizational detrimental policies to wellness.^[57] We believe this model should be adopted during advanced medical training programs in all specialties.

4.2. Study limitations

A limitation of our study is the “snapshot” results it gives, reflecting only the views at the time of the questionnaire rather than over an extended period of time which can affect the generalizability of the results. There may have been a selection bias shaped by those who responded to the questionnaire and may have had the predisposition to experiencing either more or less burnout than the overall population of trainees. Therefore, it is vital to be aware of the projecting limitations of all cross-sectional studies: Without future longitudinal data, it is not possible to get an actual idea of where the detected burnout rate comes from, i.e., to establish a true cause-and-effect relationship. Accurate assessment of trainee physician burnout is historically challenged by low response rates to studies of resident wellness and burnout, dependence on self-reporting of symptoms, and different definitions of wellness and burnout among researchers.^[58] Furthermore, our study took place a couple of months prior to Coronavirus Disease 2019 pandemic, hence, did not capture the effect the work and family-related changes had on the lives of trainees even though it is being published following the pandemic.

5. Conclusion

Burnout is clearly an issue amongst medical trainees who are parents during training. Our results identified several correlating factors and offer goals for targeted future efforts to reduce the prevalence of burnout especially among trainees starting their training and those with younger children. This can be achieved through creating a positive learning experience for trainees juggling parenting and training needs by fostering learning environments that help prevent and decrease burnout and enhance the well-being of trainees and their families. Hence, Interventions to prevent and improve the effect of burnout can be categorized into those targeting the individual and family levels, such as encouraging resilience, well-being and mindfulness courses and developing coping strategies and parenting courses and resources, and those targeting institutional levels such as, mentoring plans, and ensuring trainees are aware of available resources and possible accommodations, with the purpose being improving the working and learning atmosphere to aim at balancing training and parenting.

Author contributions

All authors made substantial contributions to the conception and design, acquisition, analysis and interpretation of data. Furthermore, they took part in drafting the article or revising it critically, gave final approval of the version to be published, and agreed to be accountable for all aspects of the work.