Offering Paid Sick and Family Leave is Not Enough: Leave Supportive Supervisor Behavior Training Improves Policy Use, Health and Work Outcomes

Abstract

Paid sick and family leave policies provide employees the right to request paid time-off for personal illness or family demands. Yet they can be underutilized and their effects show mixed results. Research has under-addressed the disconnect between formal and informal work-family support, clarifying the role of supervisors as contextual resources supporting paid leave. Integrating research on Job Demands-Resources (JD-R) theory, family supportive supervisory behaviors, and work-family policies, we evaluated training designed to increase supervisors’ Leave Supportive Supervisor Behaviors (LSSB), defined as supervisor behaviors perceived as encouraging employees to access, use, and value paid leave as a resource for managing nonwork time-off demands. We conducted a six-month field study at two U.S. universities, with random assignment to training (N=293 managers, 630 employees), and wait-list control conditions (N=252 managers, 293 employees). The intervention included voluntary on-line training, behavior self-monitoring, and a webinar. Compared to the control organization, employees in the trained organization reported higher LSSB, greater family leave use, higher job satisfaction, and stronger organizational commitment. Results showed the intervention was more beneficial for women (higher commitment), employees with childcare demands (higher job satisfaction, improved general health, higher sick leave use), and sandwiched (both child and older adult/elder) caregivers (higher job satisfaction). Post-hoc longitudinal analyses identified that the intervention increased supervisors’ role modeling behaviors, which predicted higher employee job satisfaction and LSSB. We provide future implications on the criticality of pairing supervisor training with paid leave to strengthen supervisor support as a contextual resource to improve policy use, health, and work outcomes.

“I do think as we evolve to be greater workplaces that want to provide more empathy and support, we should train managers [to communicate to employees], ‘I’m here to help shepherd you through this process’…The manager should help the employee navigate getting what they need….I think it’s educating the leader to make it safe for the employee to have a conversation that says ‘I need an accommodation’….The long game is to support people so they’ll want to work there for the long haul.”

– HR Policy Expert (Formative Research Interview #001)

Paid family and sick leave policies provide employees with the job-protected right to take partially or fully funded time-off from work for pregnancy, birth or adoption of a child, personal illness, or to care for family members (e.g., children, parents, spouses; Nandi et al., 2018). As family structures shift and societies age, paid leave has become a defining economic and social policy issue for the 21st-century (Davison & Blackburn, 2023). Paid leave can be used intermittently for a few hours (e.g., medical appointments) or all at once for days or weeks (e.g., parental leave) with duration varying depending on leave type (e.g., maternity, paternity, sickness), state, or nation. Paid leave length averages 19 weeks in OECD nations and 6–12 weeks in the U.S. (OECD Family Database, 2025). Although the United States is the only major industrialized nation without a federally-mandated paid leave policy (Hideg et al., 2018; Williamson, 2025), nearly one-third (32%) of U.S. private sector workers – approximately 46 million individuals – are now eligible under state or municipal paid leave laws (Mason & Robbins, 2026). Regardless of whether legally mandated, many U.S. employers offer paid sick (77%) and family leave (27%) voluntarily (U.S. Bureau of Labor Statistics, 2023, 2024).¹

Substantial evidence shows that paid leave is positively related to numerous health (i.e., infant mortality, post-partum recovery, breastfeeding, hospitalization length, illness exposure); work (i.e., absenteeism, turnover, stress, burnout); economic (i.e., Gross Domestic Product, labor force participation) and societal (i.e., fertility rates, poverty, gender, class, and racial equity) outcomes (Bartel et al., 2023; Goodman et al., 2022; Heymann & Sprague, 2022; Thomas et al., 2022). Yet even when available, these potential benefits are often under-realized, as many employees do not fully access, use or benefit from policies to which they are entitled (Budd & Mumford, 2006; Chen, 2016). We argue that research has yet to resolve the disconnect between the availability of formal work-family support (e.g., leave policies) and informal organizational support, namely the role of supervisors as contextual resources socially supporting employee access and use of paid leave. Studies report that employees describe paid leave supervisory conversations as being uncomfortable, career-discouraging, and conflict-laden (Buzzanell & Liu, 2007; Little et al., 2017). Requesting leave can raise fear of a flexibility stigma (Cha & Grady, 2024; Thébaud & Pedulla, 2022) and backlash affecting pay, performance and promotion evaluations (Petts et al., 2022) or even job loss (Bose et al., 2020). Moreover, many employees lack understanding of their eligibility (Elser et al., 2022) and supervisors lack understanding of how to support paid leave.

In this study, we contend that Job Demands-Resources (JD-R; Bakker et al., 2023) theory provides a strong theoretical foundation that underpins the importance of adopting organizational interventions to align formal work-family resources (leave) with informal contextual resources (e.g., supervisor support) for managing work-family demands. JD-R theory explains how the work environment affects employee well-being and work outcomes (Bakker & Demerouti, 2017) as well as the critical role of the supervisor as part of the work context. Supervisors influence how employees perceive job resources and demands, their well-being, and motivation and perceptions of agency to access job resources (Tummers & Bakker, 2021).

Integrating research on JD-R theory, family supportive supervisory behaviors (Hammer et al., 2011, 2019; Kossek et al., 2018, 2024), and work-family policies, the current research designed and evaluated a theoretically-grounded organizational training intervention that aimed to increase Leave Supportive Supervisor Behaviors (LSSB), and health and work outcomes. LSSB is defined as supervisor behaviors perceived as encouraging employees to access, use, and value paid leave as a resource for managing nonwork time-off demands. JD- R theory would suggest that supervisors influence how employees perceive job resources, including their motivation to access and benefit from them. Given the many informal social support barriers to paid leave (Hipp et al., 2022), we believe that LSSB is the mechanism by which employees are likely to perceive leaves as a viable valued resource. We assume that supervisors exhibiting emotional, instrumental and other supportive behaviors for paid leaves is critical to enhance their perceived value as contextual resource. Previous research and policy have focused more on adopting, funding, or lengthening paid leave than on enhancing contextual support for leave access, use and effectiveness (Elser et al., 2022); they also under-emphasize the supervisor role as most firms rely only on the human resource department to manage leave support. While research continues to document problems involving a decoupling mismatch between paid leave formal availability and employee perceptions of informal support for leave access and use, we lack understanding and solutions on how to enhance leave effectiveness as a resource. The intervention was designed to enhance supervisors’ awareness, motivation, and skills for supporting leave and improving employees’ LSSB perceptions in the organization context, to positively influence work and health outcomes. We argue that employees in a context with trained supervisors will be more likely to perceive paid leave as an effective resource to support family needs or personal illness compared to those employees in a context where they do not perceive supervisors as supportive or aware of how to encourage policy use.

Study Specific Goals, Overview and Contributions

In this field study (see Figure 1), we examine whether a supervisor training intervention designed to increase paid Leave Supportive Supervisor Behaviors (LSSB) is associated with higher employee perceptions of LSSB, more positive work attitudes (affective organizational commitment, job satisfaction), improved health outcomes (lower psychological distress, higher physical health), and increased paid leave policy use in an organization with paid family and sick leave, compared to an organization with paid leave but without supervisor support training. This design enables an evaluation of the effects of combining the availability of a formal paid leave policy with training fostering informal supervisory support versus having a formal paid leave policy offered without such support.² This identifies how organizations, leaders, scholars, and policymakers can improve paid leave use and effectiveness.

Figure 1. Study Model: Organizational Leave Supportive Supervisor Behaviors (LSSB) Training Intervention Effects on Employee Policy Use, Work Attitudes, and Health Outcomes.

Second, we extend JD-R theory by understanding how integrating formal and informal work-family supports can enhance the effectiveness of paid leave through the theoretical mechanism of increased perceptions of LSSB. We assume that job resources (training to motivate supervisors to increase paid leave supportive behaviors) can help create a culturally supportive context for employees to not only manage job demands as traditionally studied, but to also better leverage job resources for managing the demands of the work-family interface.

Theoretical Background and Hypotheses Development

JD-R theory posits that job characteristics can be classified as job resources or job demands, which are causally linked to employee health and job outcomes (Demerouti et al., 2001). Another assumption is that work resources (e.g., supervisors, policies) and individual resources interact to improve job well-being and health outcomes (Bakker et al., 2023).

Research suggests supervisors are key actors shaping perceptions of resources, including workplace contextual social support for the work-family interface (Kossek et al., 2011; Hammer et al., 2011), and in accessing resources and using leaves specifically (Davison et al., 2023; Elser et al., 2022; Goodman et al., 2022). We know supervisors act as gatekeepers who shape employees’ sense of legitimacy and safety using human resource policies (Kehoe & Han, 2020). Yet many supervisors may not be motivated to support paid leave; they may see access and use as a private or uncomfortable matter, and a headache to cover workload, scheduling, and human resource department coordination (Little et al., 2017). Even when legally mandated, supervisors can lack understanding of leave policy eligibility and administrative requirements, resulting in noncompliance (Goodman et al., 2020). This shortfall in supervisor support is not just a U.S. issue but can occur internationally, even in countries (e.g., Germany, Japan) with strong legal mandates (Hipp et al., 2022; Takahashi et al., 2014). Yet most countries assume that offering paid leave as a formal resource is sufficient to ensure they will be seen as accessible and used to manage the demands of the work-family interface and overlook supervisor supportive behaviors.

Family supportive supervisor behaviors (FSSB) include role modeling, instrumental, emotional, and creative management support (Hammer et al., 2009). They can be conceptualized as an informal social resource that influence how employees perceive support for and accessing resources for managing the work-family interface, which influences nonwork and work outcomes (ten Brummelhuis & Bakker, 2012). Over time, JD-R theory has evolved to encompass multi-level perspectives (Bakker, 2015) and recognizes that individuals and groups actively shape job demands and resources within dynamic work contexts (Apenbrink & Kuhlmann, 2025). Consistent with this approach, we adopt a multi-level view (Figure 1) to consider how context and multiple actors (i.e., supervisors and employees) shape job demands and resources.

Leave Supportive Supervisor Behaviors (LSSB) – Measurement and Training Development

We designed a scale and corresponding training intervention to enhance LSSB, defined as supervisor behaviors perceived as encouraging employees to access, use, and value paid leave as a resource for managing nonwork time-off demands. Drawing on prior research on FSSB (Hammer et al., 2009), a literature review on workplace barriers to paid leave effectiveness (e.g., Baird & Reynolds, 2004), and interviews with HR policy experts (see Appendix 1 Background, for details), we identified six supervisor behaviors central to supporting paid family and sick leave: role modeling leave use, educating and increasing access to leave, emotional support, fostering a dual agenda, psychological safety for leave, and respecting time off during leave. Table 1 provides definitions, and scale development details can be found in the Method, and Supplement Table 2 and Supplement Appendix 1).

Table 1.

Leave Supportive Supervisor Behaviors (LSSB) Summary for Developing Training Intervention and Measure

	Type of Supervisor Support	Definition of Support	Example Support Behaviors
L	Listen and Give Emotional Support	Behaviors that make an employee feel comfortable sharing their needs for time off for family and sick leaves	• Actively listen to employee’s requests by reading their body language, and paying attention to the tone of their voice • Demonstrate empathy by communicating genuine concern about what your employee is going through and encourage them to use available leave when they have the need
E	Educate and Increase Policy Access	Be knowledgeable about policies and helping employees access their options	• Educate yourself, your employees, or other managers on leave policies and the negative effects of not taking leave when needed • Help employees understand how much control over timing and length of leaves they have
A	Address Negative Perceptions Through Psychological Safety	Reassuring the employee and creating a supportive work environment surrounding leaves	• Reassure the employee that they will not be penalized for being away and reduce stigma by making employee’s less afraid to take the time off they need and normalize leave requests • Be aware of how leave takers are discussed among colleagues and clients while away, and champion them when needed
D	Dual Agenda for Fostering Creative Management	Creatively organizes work to support leaves in a way that supports dual values of employer productivity and employee time off	• Develop back-up systems like standardized paid family and sick leave policies across teams, cross-train, staff preparation, and being proactive with policies to prepare for times when a worker is taking leave • Communicate to other managers how you are being responsive to employee requests and partner with them to support each other
E	Enhancing The Employee Experience Through Respecting Time Off	This behavior is focused on providing support during the leave	• Ask employee communication preferences when they are taking leave knowing that many employees may not be comfortable being contacted while they are away particularly for routine matters • Stay connected to your employee as agreed upon during their leave, for example, ask them how they would like to receive work updates while they are away
R	Role Modeling Leave Use	Actions that show how you are taking care of your own work-life and time off challenges	• Express belief in taking time off for family and sickness and sharing your own success stories and benefits you have experienced from taking time off • Talk about the importance of self-care and when they have taken time off for illness or to care for a child or older adult

The training was intentionally developed to be brief, low-cost, and scalable, consistent with prior evidence-based supportive supervisor training programs (e.g., Hammer et al., 2011, 2019; Kossek et al., 2018, 2024). The intervention followed Walton’s (2014) principles of wise interventions, which demonstrate that concise, precisely targeted experiences can shift individuals’ “psychological reality” and meaning systems in ways that address persistent social problems. Such brief interventions are designed not to overhaul organizational systems but to activate enduring cognitive and behavioral shifts in work context. Implementing supervisor training not only signals organizational commitment to work-family support but also enhances supervisors’ motivation, skills, and confidence to engage in leave-supportive behaviors.

Hypothesis 1: Training supervisors to support paid sick and family leave increases employee perceptions of LSSB compared to employees in the control organization where the supervisors were not trained.

Main Effect Links Between Training, Policy Use, and Job and Health Outcomes

Prior research suggests that supervisory cultures can inhibit leave utilization, particularly in occupations with permeable boundaries and high productivity pressures, where employees often report continuing to work during their leave time (Huppatz et al., 2018). A lack of supervisor knowledge and awareness regarding paid leave policies may represent a key barrier that a targeted training intervention can address. For example, in an evaluation of San Francisco’s first mandated fully paid parental leave policy, Goodman et al. (2020) found that nearly half of covered employers were noncompliant with the ordinance, and among those, almost half struggled to understand their legal responsibilities.

Yet many employees who would benefit from paid leave policies choose not to use them due to concerns about burdening coworkers, harming workplace relationships, or experiencing stigma and negative career repercussions (McNamara et al., 2012; Shauman et al., 2018; Smith et al., 2019; Thébaud & Pedulla, 2022). Consistent with evidence showing a persistent decoupling between policy availability and perceived usability (Eaton, 2003), these barriers highlight the importance of contextual resources that shape how policies are enacted and experienced.

We therefore assume that employees in organizations offering supervisor training will experience a stronger and more supportive human resource context (Bowen & Ostroff, 2004) that reduces the gap between policy availability and actual use. In such environments, supervisors are more motivated and capable of facilitating access to leave, and employees in such contexts will be more likely to perceive leave as less stigmatized, more usable, and personally accessible. As a result, they should feel more comfortable and motivated to request and use leave when needed.

Hypothesis 2: Training supervisors to support paid leave increases employee paid sick and family leave use compared to employees in the control organization where the supervisors were not trained.

Within JD-R, paid leave functions as an organizational resource that enables employees to recover and manage nonwork demands, thereby preventing strain and promoting positive outcomes. Employees whose supervisors receive training are expected to experience higher LSSB, reflecting more personalized attention to their unique circumstances for leave. This enhanced supervisor support should, in turn, foster greater job satisfaction and affective organizational commitment.

From a JD-R perspective, supervisors play a pivotal role in allocating and managing job demands and resources. When leaders are trained to support leave, they can distribute resources more effectively across the workforce. Supportive leaders who balance demands and resources create environments where employees remain healthy, motivated, and productive (Schaufeli, 2015), conditions consistently associated with higher satisfaction and commitment.

Hypothesis 3: Training supervisors on LSSB increases employee job satisfaction and affective organizational commitment compared to employees in the control organization where the supervisors were not trained.

We propose that employees in organizations where supervisors are trained to enact LSSB will experience better mental health (lower psychological distress) and better physical health than employees in organizations without such training. Within the JD-R framework, supervisor training enhances both formal resources (paid leave availability) and informal resources (supervisory support) in the work context to manage work-nonwork demands. When employees know they can access leave and perceive supervisor encouragement to use it, they experience less anxiety (Schaufeli, 2015) and are more likely to engage in proactive health behaviors, such as seeking medical care or taking adequate recovery time. Empirical evidence supports these effects: using 2019–2020 Longitudinal National Health Interview Survey data, Asfaw (2024) found that workers with paid sick leave were significantly less likely to report daily depression, anxiety, and both depression and anxiety, than those without such access. Paid maternity leave likewise reduces postpartum depression, maternal rehospitalizations, and related health risks (Van Niel et al., 2020). Supervisors who avoid pressuring employees to return to work prematurely enable full recovery and prevent longer-term negative health consequences.

Organizations that train supervisors to actively support leave foster climates where employees feel safe prioritizing their mental and physical health, which has been linked to lower prenatal stress, improved postpartum health, and shorter recovery times (Jones et al., 2021).

Hypothesis 4: Training supervisors on LSSB reduces employee perceived distress and improves physical health compared to employees in the control organization where the supervisors were not trained.

Moderating Effects: Training Benefits for Women and Child and Adult Caregivers

Drawing on JD-R theory, we anticipate that the intervention will be especially beneficial for employees who have traditionally had access to fewer psychological and social resources in their organizational context to manage the work-nonwork interface but face higher family demands requiring paid leave. This expectation aligns with prior intervention research (Brady et al., 2021; Kossek et al., 2019; Hammer et al., 2011, 2019; Mohr et al., 2024), which consistently demonstrates moderated effects wherein employees with the greatest need at baseline (pre-intervention) derive the strongest improvements in health and work outcomes following implementation of supportive workplace interventions.

This “need” includes employees with higher childcare and older adult care³ responsibilities, who are disproportionately women and often manage overlapping caregiving roles while working (U.S. Bureau of Labor Statistics, 2023b). Studies of nursing home employees further illustrate this pattern: “double” and “triple-duty” caregivers, who are predominantly women who simultaneously care for children, older relatives, and residents needing living assistance, reported reduced psychological distress when working in contexts with trained, family-supportive supervisors (Kossek et al., 2019). Accordingly, caregivers and women with trained supervisors may perceive paid leave as a more usable and accessible resource. By enhancing access to supportive resources, such environments are expected to promote greater work, health, and policy use among employees with higher care demands.

Hypothesis 5: Gender moderates the impact of training LSSB on employee perceptions of supervisor supportive behaviors for paid leave on work outcomes (job satisfaction, organizational affective commitment), health outcomes (psychological distress, physical health), and paid leave use policies, such that women will have more positive outcomes compared to men.

Hypothesis 6: (6a) Childcare (6b) elder/older adult caregiving and (6c) sandwiched care responsibility moderates the impact of training LSSB on supervisor supportive behaviors for paid leave, work outcomes (job satisfaction, affective organizational commitment), health outcomes (psychological distress, physical health), and paid leave policy use, such that those with higher child, elder/adult, and sandwiched care (both) responsibilities will have more positive outcomes compared to those with lower caregiving responsibilities.

Method

Research Sites and Participants

We conducted the training intervention in two nationally-ranked public research universities located in two U.S. states without state-level paid leave policy mandates, interested in improving their work-life benefits. University 1 is in the Southwest with a medical center. University 2 is in the Midwest with the most popular majors in nursing and other health professions, business, engineering, and social sciences. Although not legally required, each University offered paid family and sick leave to its faculty and staff. In University 1, all employees receive 1 month of childcare leave for birth, adoption or care, and 24 hours of sick leave each year. University 2 offers six weeks of paid parental leave, and ten hours of leave each month accrued for each month of service, with a pro-rate for 9- or 10-month contracts⁴.

Study participants were managers (defined as an individual in a supervisory role in a formal capacity and someone with direct reports) and their direct reports (referred to as employees in this paper). Demographic characteristics of the managers and employees by university appear in Supplement Table 1.

For both universities, participants were voluntarily invited to complete a baseline survey and two follow-up surveys approximately 3 and 6 months after the baseline survey (referred to as Times 1, 2, and 3). All employee participants, whether they completed the Time 1 baseline survey, were invited to participate in follow-up surveys. 630 employees at University 1 (of the 1899 employees invited to participate (33%) and 293 employees at University 2 (of the 811 employees invited to participate, (36%) completed at least one survey. See Supplement Figure 1 for an overview of recruitment procedures across the two organizations.

Study Design

This 6-month field study utilized a two-condition pre- and multiple-posttest cluster group-randomized design where the two universities were randomly assigned to either a training condition where supervisors received the LSSB training or a wait-list control condition where supervisors did not receive training until after the final data collection period for the study (See Figure 2 for a study design overview). Allocating organizations to the contrasting study conditions was used to minimize the effects of contamination of training information from trained supervisors to untrained supervisors in the same organization, which helps prevent a fair appraisal of any training benefits (for a similar design randomizing organizations on leadership training and health and work outcomes (see Hauth, Peiró, & Mesa, 2022).

Figure 2. Overview of Field Study Design.

Data Transparency and Open Science Statement

This study Family Supportive Supervisor Training and Workplace Assessment was not a preregistered study but was registered as a clinical trial with the ID NCT05080426 at https://clinicaltrials.gov/. It had ongoing external review by a data safety monitoring board as a condition of the U.S. National Institutes of Health Grant #2R42AG060347-02A1 NIA grant, including approval by Purdue University’s IRB Institutional Research Board (2020–1687). Consistent with open science practices, we provide sufficient detail and materials to enable independent replication of the study’s design, measures, and analyses including analytic syntax, and de-identified data, which are publicly available as additional online material at the Open Science Framework. The training files also can be accessed at Purdue University. This is an original dataset and the first publication from these data. For any other study-related information, please contact the corresponding author.

All analyses were conducted using Mplus Version 9 (Muthén & Muthén), along with IBM SPSS Statistics for data preparation and preliminary analyses. The authors confirm that this manuscript complies with the Journal of Applied Psychology methodological reporting checklist.

Development and Delivery of LSSB Training Intervention

The intervention was designed to increase supervisors' Leave Supportive Supervisor Behaviors (LSSB) outlined in Table 1. The intervention included: 1) two on-line training modules (self-paced; about an hour or less each),⁵ 2) post-training behavioral self-monitoring (with goal setting activities and reminders to record supportive behaviors to foster training job transfer), and 3) a 90-minute interactive organization wide webinar to discuss training concepts. These varied voluntary activities gave managers flexibility to participate in whatever learning component for which they had time and interest. Managers were told they would receive a certificate for participating in training and a personalized feedback report. The intervention was developed by two work-family scholars (and paper co-authors) who were highly experienced in developing, implementing, and evaluating evidence-based supervisor support interventions in organizational field research.

The online training software included active listening and participation components (e.g., matching activities; quizzes), and examples of situations managers may encounter, coupled with examples of supportive strategies to utilize. The first module, Supportive Supervision and Management for Family and Personal Life. was a module on the general supervisor role in family support. Building on earlier research on family supportive supervisor behaviors, it included updated content on the criticality of the supervisor role in supporting family and personal life support, its benefits, and examples of growing nonwork support needs. The second module, Leaders and Leaves: Supporting Employee Use of Paid Family and Sick Leave Policies, focused on LSSB (see Supplement Table 2, Figures 2 and 3 for illustrative content).

Figure 3. Moderated Intervention Effects: Gender, Childcare, Adult Care and Sandwiched Care (both child and older adult care) Responsibility.

Procedure

Each organization designated a primary point of contact (POC) for the research team. The POC developed a list of all managers and employees who would be invited to participate in the study and helped facilitate recruitment. Each POC provided invitees’ names, work email addresses, role designations (i.e., manager or employee), and identified each invitee’s manager to allow for linking in subsequent data analyses. These details were uploaded into a customized online learning management system (LMS), which is a software application that has the capability to automate the delivery and tracking of educational programs. This LMS system sent out recruitment emails to managers and employees that provided participants with a link to complete online surveys and (for managers) to participate in training activities. After Time 1 recruitment, intervention training became available for managers at University 1. These managers were sent five recruitment emails about the training over one month as reminders. Reminders about behavior tracking were sent to managers who had completed both training modules, After the training closed, all managers (even those who did not participate in the online training) were sent an email by an internal leadership development department inviting them to participate in a webinar, approximately one month after the training closed.

The webinar was led by the experts who developed the intervention. It consisted of a review of the supportive strategies learned in the training followed by an interactive discussion of how participants implemented the strategies. Following the webinar, managers and employees at both universities were invited to participate in the follow-up data collection surveys (even if they did not complete the baseline survey). At University 1, 369 managers were invited to participate in the three intervention training components, of which 174 managers (47%) completed at least one component. Of the 174 managers who completed the training, 30 managers completed all three components (17.2%), 46 completed two components (26.4%), and 98 completed only one component of the training (56.3%).

Measures

For all scales, employees reported their agreement with the items using a 5-point rating scale (1 = strongly disagree; 5 = strongly agree). See Table 2 for the means, standard deviations, and correlations of all variables. See Supplement Table 3 for descriptive statistics by condition across the two universities at T1 (baseline), Time 2 and Time 3.

Table 2.

Means, Standard Deviations, and Correlations of Study Variables

		Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28
1	Intervention	0.68	0.47	--
2	Leave Sup. Support. Beh (LSSB) (T1)	4.01	0.84	0.05	--
3	Leave Sup. Support Beh. (LSSB) (T2)	4.05	0.83	0.04	0.78	--
4	Leave Sup. Support Beh (LSSB) (T3)	4.02	0.88	0.11	0.76	0.79	--
5	Commitment (T1)	3.67	0.97	0.10	0.33	0.24	0.26	--
6	Commitment (T2)	3.77	0.95	0.17	0.24	0.25	0.24	0.62	--
7	Commitment (T3)	3.67	0.99	0.19	0.36	0.27	0.38	0.59	0.73	--
8	Job Sat. (T1)	4.13	0.80	0.05	0.43	0.33	0.38	0.44	0.35	0.41	--
9	Job Sat. (T2)	4.14	0.81	0.08	0.40	0.49	0.46	0.32	0.42	0.45	0.69	--
10	Job Sat. (T3)	4.08	0.84	0.07	0.42	0.42	0.51	0.38	0.37	0.51	0.69	0.78	--
11	Distress (T1)	1.90	0.79	−0.02	−0.29	−0.24	−0.22	−0.15	−0.20	−0.18	−0.37	−0.40	−0.40	--
12	Distress (T2)	1.86	0.77	−0.03	−0.22	−0.32	−0.33	−0.12	−0.23	−0.29	−0.42	−0.44	−0.51	0.76	--
13	Distress (T3)	1.82	0.76	0.01	−0.29	−0.30	−0.31	−0.13	−0.25	−0.24	−0.34	−0.45	−0.39	0.70	0.83	--
14	Health (T1)	2.87	0.69	0.01	0.11	0.18	0.18	0.11	0.04	0.03	0.13	0.16	0.20	−0.32	−0.32	−0.21	--
15	Health (T2)	2.85	0.72	0.01	0.12	0.16	0.17	0.14	0.16	0.22	0.20	0.19	0.28	−0.33	−0.38	−0.36	0.71	--
16	Health (T3)	2.86	0.71	−0.03	0.13	0.14	0.17	0.07	0.11	0.11	0.15	0.20	0.22	−0.35	−0.34	−0.34	0.77	0.76	--
17	Fam. Leave Use (T1)	2.14	12.33	0.06	−0.02	−0.16	−0.07	0.05	−0.09	−0.05	−0.04	−0.20	−0.08	0.04	0.16	0.11	−0.02	−0.03	−0.06	--
18	Fam Leave. Use (T2)	1.96	8.89	0.04	−0.15	−0.08	−0.12	−0.01	−0.01	−0.08	−0.02	−0.07	−0.08	0.10	0.07	0.18	−0.11	−0.06	−0.10	0.18	--
19	Fam. Leave Use (T3)	1.94	8.15	0.13	−0.02	−0.06	−0.01	−0.09	0.04	0.02	−0.04	−0.10	−0.03	0.03	0.08	0.05	−0.04	−0.01	−0.03	0.14	0.85	--
20	Sick Leave Use (T1)	2.83	5.67	−0.07	−0.02	0.02	0.02	0.00	−0.06	−0.01	−0.06	−0.08	−0.07	0.04	0.12	0.05	−0.11	−0.11	−0.13	0.12	0.06	0.10	--
21	Sick Leave Use (T2)	2.93	6.77	−0.14	0.01	−0.01	0.04	−0.01	−0.04	−0.02	−0.03	−0.03	−0.03	0.03	0.08	0.07	−0.10	−0.14	−0.09	0.11	0.12	0.04	0.55	--
22	Sick Leave Use (T3)	3.09	7.81	−0.05	−0.05	−0.07	−0.02	−0.16	−0.12	−0.08	−0.07	−0.16	−0.06	0.07	0.18	0.05	−0.11	−0.18	−0.11	0.10	0.16	0.23	0.22	0.41	--
23	Age (T1)	46.11	11.24	−0.17	−0.02	0.01	0.02	0.09	0.14	0.08	0.10	0.17	0.11	−0.22	−0.23	−0.18	0.05	0.04	0.02	0.00	−0.03	0.00	0.01	0.08	−0.03	--
24	Kids (T1)	0.45	0.50	0.08	−0.06	−0.04	−0.03	0.05	0.03	−0.01	0.02	−0.06	−0.12	0.05	0.08	0.07	−0.06	−0.07	−0.09	0.06	0.09	0.04	0.05	0.01	−0.05	−0.16	--
25	Education (T1)	5.24	0.90	−0.41	−0.01	0.03	−0.05	−0.07	−0.04	−0.12	−0.11	−0.08	−0.08	0.11	0.20	0.13	0.07	0.04	0.12	0.02	0.05	−0.05	−0.02	0.06	0.01	−0.02	−0.03	--
26	Gender (T1)	0.74	0.44	0.07	−0.08	−0.06	−0.08	0.03	0.06	0.10	0.01	−0.02	0.01	−0.03	0.03	−0.01	0.00	0.01	0.01	0.03	0.02	0.04	0.06	0.00	0.03	−0.07	0.03	0.01	--
27	Childcare responsibility (T1)	3.07	1.38	0.03	−0.09	−0.21	−0.17	0.02	−0.15	−0.07	0.12	−0.11	−0.10	0.07	0.22	0.16	−0.12	−0.14	−0.13	0.06	0.10	0.09	0.12	0.06	0.05	−0.27	0.31	−0.05	0.13	--
28	Eldercare responsibility (T1)	2.81	1.30	0.02	0.01	−0.09	−0.09	0.00	0.04	−0.06	0.03	0.00	−0.04	0.09	0.12	0.07	−0.03	0.01	0.00	0.08	−0.04	0.02	0.08	0.00	0.23	0.00	0.01	−0.12	0.09	0.47	--
29	Sandwiched responsibility (T1)	5.94	2.27	0.07	0.08	−0.05	−0.05	0.04	0.12	0.10	0.11	−0.07	−0.05	0.03	0.25	0.13	−0.06	0.00	−0.06	0.12	−0.12	0.11	0.15	0.03	0.16	−0.35	0.30	−0.07	0.09	0.88	0.84

Note. Significant correlations (p < .05) are bold.

Leave Supportive Supervisor Behaviors (LSSB)

To measure LSSB, formative research, as noted in the literature review, including background interviews with HR experts (see Appendix 1 below) for interview protocol and sample findings) was conducted to generate content to design the intervention and identify six leader behavior dimensions defined in Table 1 with two items representing each dimension, for a total of 12 items. Alpha reliability was .96 at baseline, .95 at Time 2, and .96 at Time 3. We used longitudinal data collection from 694 employees to validate the new measure. Results of multilevel confirmatory factor analyses found six subscales: listening and giving emotional support, role modeling, educating and increasing access, dual agenda, psychological safety for taking leave, and respecting time off. This analysis best supported a second-order factor where each of these subdimensions mapped onto a higher order factor of LSSB. For more methodological details, see Supplement Table 2 for scale items and CFA results and Supplement Appendix 1 Additional Information on Background Interviews and LSSB Scale Development and Supplement Appendix 2 Scale Validation Results for validity information.

Employee Job Outcomes

Job satisfaction was measured using 3 items from Cammann and colleagues (1983) scale. Items included the following: “In general, I like working at my job”, “In general, I am satisfied with my job”, and “I am generally satisfied with the kind of work I do in this job”. The reliability for this scale was .91 at baseline and Time 2, and .92 at Time 3.

Organizational commitment was measured using one item from Allen and Meyer’s (1990) three item organizational commitment scale: “I feel emotionally attached to my organization”.

Employee Health, Caregiving Responsibilities, and Paid Leave Use Outcomes

Psychological distress was measured using Kessler and colleagues’ (2003) 6-item scale (e.g., item: “During the past 30 days, how much of the time did you feel so sad nothing could cheer you up?”). The reliability for the psychological distress scale was .88 at baseline, .89 at Time 2, and .90 at Time 3.

General health was measured using one item from (Ware & Sherbourne,1992): “In general, how would you rate your health?”

Use of Paid Leave was measured using two items developed by the study authors: “How many days of sick leave have you used in the past 6 months?” and “How many days of family leave have you used in the past 6 months?”

For individuals who reported childcare and/or elder/older adult care responsibilities, the Extent of Caregiving responsibilities were assessed using a continuous measure separately for childcare and older adult care and then combined to measure sandwiched care using the stem (1= None of the time, 5 = All of the time). The childcare scale was derived from Kossek (1990) and consisted of three items: (1) To what extent are you personally responsible for organizing and managing the supervision and care of your children under 19 while you are at work?, (2) To what extent are managing problems with the quality of supervision and care of your children a concern for you?, and (3) To what extent are managing problems with the dependability of supervision and care of your children a concern for you?

We measured the extent of responsibility for managing older adult/eldercare demands with the same 3 items but replaced childcare with elder/ adult care terminology: Sample item (If caring for elders…to what extent are managing problems with the quality of supervision and care of your older adult a concern for you? Sandwiched care was assessed by combining elder and childcare responsibility. Reliability for managing childcare was .91, managing elder or adult care was .92, and sandwiched care was .93 at baseline.

Analytic Approach

We used two-level analysis of covariance (ANCOVA) models to evaluate the main effects of training on employee outcomes (Bodner & Bliese, 2018) using a training condition indicator variable and controlling for baseline values of the employee outcomes. Moderated training effects were evaluated by adding an interaction term involving the training condition indicator and the moderator variable; continuous moderator variables were grand-mean centered in these statistical models. An exploration of intraclass correlations (ICCs) across the study outcomes at follow-up based on the nesting of employees within supervisors indicated small to modest ICCs ranging from .01 to .12; to be conservative, we therefore accounted for the nesting of employees in supervisors in our models with a supervisor-level random effect.

To minimize bias due to differences in employees across the two universities, we tested for differences in study and demographic variables at baseline. Significant differences were found across universities for employee age, number of kids, and education (see preliminary analyses below). Thus, we controlled these variables assessed at baseline in the models for all hypothesis testing. Although these control variables did not alter study results, their inclusion in the model provides greater assurance to minimize any preexisting organizational influences or differences in field study findings.

Our use of a training condition indicator variable at the university level, with a variable value held in common for all employees in that university, aligns with the recommended intent-to-treat (ITT) approach when evaluating intervention effectiveness (Gupta, 2011). This approach includes employees in the treatment condition regardless of whether their managers received or completed the LSSB intervention. As such, this approach is considered conservative when estimating intervention effects, reduces selection effect biases that would occur if managers self-selected to participate or not participate in the training, and provides a realistic assessment of the training intervention’s impact in organizations where manager training is not mandatory.

Separate models were used to evaluate main and moderated training program effects on the study outcomes at each of the follow-up assessment periods. To evaluate potential differences in intervention effects on the study outcomes across the two follow-up assessment periods, a model comparison approach was implemented. These model comparisons compare the fit of unconstrained (i.e., permitting the training effects to be different over time) and constrained (i.e., constraining the training effects to be equal over time) models where a significant difference in model fit indicates different training effects over time. This approach avoids overinterpretation of effects observed in training over time differences (e.g., p = .04 and p = .11 for the two time periods) and allows for broader statistical conclusions about the effectiveness of the training intervention (see Bodner, 2018; Brady et al., 2021 and see Supplement Appendix 5 Model Comparisons for information on model comparisons of all results below).

Missing data are common in studies with repeated assessments over time. To estimate and test for main and moderated training effects on the study outcomes, all analyses were conducted in Mplus V8 (Muthén & Muthén, 2017) using the full information maximum likelihood approach. This approach provides unbiased estimates of training effects in the presence of missing data and adjusts upwards the inferential uncertainty due to missing observations when the Missing At Random (MAR; Little & Rubin, 1987) assumption holds; our inclusion of baseline values for the outcome variables in our statistical models increases the plausibility of the MAR assumption (Enders, 2010). Below we elaborate on missing data prevalence and manager training participation⁶.

While our primary analytical approach to evaluate overall training program effectiveness used Intent-to-Treat, we conducted additional supplement analyses within University 1 to evaluate whether employee outcomes differed across different patterns of manager participation in the training (e.g., an intervention dose-response analysis; Supplement Appendix 3: Attritors Analysis). No significant differences were revealed modeling manager training status on any of the primary study employee outcomes. The lack of significant differences imply that no one training component or subsets of the available training components were more critical for evaluating the overall effectiveness of the training program in subsequent analyses under Intent-to-Treat presented below.

Results

Manipulation Check of LSSB Training Effects: Manager Learned Behaviors

To better address additional fidelity checks to demonstrate the effectiveness of the manipulation of the manager training, we assessed primary intervention effects on whether trained managers (i.e., within-subjects tests) reported greater perceptions of eliciting support for paid leave to their employee’s post-intervention. Main effects of the intervention on manager outcomes indicated a significant increase on self-reported LSSB policies at Time 2, three months post-baseline (b = .307, SE = .054, p < .001). No significant increase was revealed in manager perceptions of themselves eliciting LSSB at Time 3 (b = −.069, SE = .050, p = .168). Thus, we consider this significant main effect as a manipulation check that managers felt that they learned this specific type of support 3-months post-baseline. See Supplement Appendix 4: Manipulation of manager effects for further manipulation check and fidelity analyses and information. ⁷

Main Intervention Effects

Results of main intervention effects on LSSB, paid leave usage, workplace attitudes, and health can be viewed in Tables 3 and Supplement Tables 4–9.

Table 3.

Main Effects of LSSB Intervention on Use of Paid Leave, Workplace Attitudes, and Health Outcomes

	Leave Supportive Supervisor Behaviors (LSSB)			Use of Paid Family Leave			Use of Paid Sick Leave			Psychological Distress			Physical Health			Organizational Commitment			Job Satisfaction
3-months
Variable	B	SE	95% CI	b	SE	95% CI	b	SE	95% CI	B	SE	95% CI	b	SE	95% CI	b	SE	95% CI	b	SE	95% CI
Intercept	3.99***	0.03	[3.92, 4.05]	1.93***	0.37	[1.21, 2.65]	2.98***	0.29	[2.41, 3.55]	1.84***	0.03	[1.79, 1.90]	2.84***	0.03	[2.79, 2.90]	3.74***	0.04	[3.66, 3.81]	4.12***	0.03	[4.05, 4.18]
Intervention	0.05	0.07	[−.09, .19]	.85***	0.09	[.68, 1.02]	−1.01	0.64	[−2.26, .25]	−0.02	0.06	[−.13, .10]	0	0.28	[−.12, .17]	.22**	0.09	[.06, .39]	.17*	0.07	[.03, .31]
Total Model R2	0.58			0.32			0.27			0.58			0.52			0.41			0.5
	Leave Supportive Supervisor Behaviors (LSSB)			Use of Paid Family Leave			Use of Paid Sick Leave			Psychological Distress			Physical Health			Organizational Commitment			Job Satisfaction
6-months
Variable	B	SE	95% CI	b	SE	95% CI	b	SE	95% CI	B	SE	95% CI	b	SE	95% CI	b	SE	95% CI	b	SE	95% CI
Intercept	4.03***	0.03	[3.97, 4.09]	2.35***	0.37	[1.62, 3.08]	3.03***	0.39	[2.26, 3.80]	1.82***	0.03	[1.77, 1.88]	2.87***	0.03	[2.82, 2.92]	3.67***	0.04	[3.59, 3.75]	4.10***	0.03	[4.03, 4.16]
Intervention	.15*	0.07	[.02, .28]	2.23***	0.03	[2.17, 2.29]	−0.6	0.8	[−2.17, 1.0]	0.06	0.06	[−.07, .18]	−0.01	0.05	[−.11, .10]	.24**	0.09	[.06, .41]	0.1	0.07	[−.03, .24]
Total Model R2	0.58			0.08			0.05			0.53			0.6			0.39			0.51

Note. All models controlled for baseline levels of the outcome variable, age, education, and childcare role (0 = no, 1 = yes). All estimates represent unstandardized values.

^*p < .05.

^**p < .01.

^***p < .001.

LSSB & Leave Use Outcomes

LSSB.

There was no significant main effect of the intervention on employee reports of LSSB at Time 2 (b = .05, SE = .07, p = .446, pseudo ΔR² = .00; d = .10). However, the intervention significantly increased employee reports of LSSB at Time 3 (b = .15, SE = .07, p = .027, pseudo ΔR² = .01; d = .27). Hypothesis 1 was partially supported.

Family Leave Use.

The intervention significantly increased employee reports of paid family leave usage at Time 2 (b = .85, SE = .09, p < .001, pseudo ΔR² = .01; d = .12) and at Time 3 (b = 2.23, SE = .03, p < .001, pseudo ΔR² = .01; d = .28).

Sick Leave Use.

No significant intervention effects were revealed for employee reports of paid sick leave use at Time 2 (b = -1.01, SE = .64, p = .116, pseudo ΔR² = .00; d = −.17) or Time 3 (b = −.60, SE = .80, p = .459, pseudo ΔR² = .00; d = −.08). Taken together, these findings indicate that Hypothesis 2 was partially supported.

Work Outcomes

Job Satisfaction.

The intervention significantly increased employee reports of job satisfaction at Time 2 (b = .17, SE = .07, p = .020, pseudo ΔR² = .02; d = .29). However, there was no significant main effect of the intervention for employee reports of job satisfaction at Time 3 (b = .10, SE = .07, p = .140, pseudo ΔR² = .00; d = .17).

Affective Organizational Commitment.

The intervention significantly increased employee reports of affective organizational commitment at Time 2 (b = .22, SE = .09, p = .009, pseudo ΔR² = .02; d = .30), and Time 3 (b = .24, SE = .09, p = .008, pseudo ΔR² = .01; d = .30). Hypothesis 3 was supported at Time 2 but not Time 3.

Health Outcomes

Psychological Distress.

There was no significant intervention effect on employee reports of psychological distress at Time 2 (b = −.02, SE = .06, p = .797, pseudo ΔR² = .00; d = −.03) or Time 3 (b = .06, SE = .06, p = .372, pseudo ΔR² = .00; d = .11).

General Health.

There was no significant intervention effect on employee reports of general health at Time 2 (b = −.00, SE = .06, p = .991, pseudo ΔR² = .00; d = −.01) or Time 3 (b = −.01, SE = .05, p = .920, pseudo ΔR² = .00; d = −.01). Thus, Hypothesis 4 was not supported.

Moderated Intervention Effects

See Table 4 for moderated intervention effect results.

Table 4.

Moderated Intervention Effects of Gender, Childcare, Eldercare, and Sandwiched Care (both child and older adult care)

	Use of Family Leave		Use of Sick Leave		Leave Supportive Supervisor Behaviors (LSSB)		Psychological		General		Job		Organizational
3-months							Distress		Health		Satisfaction		Commitment
Variable	b	SE	b	SE	B	SE	b	SE	B	SE	b	SE	b	SE
Intercept Gender	1.92***	0.37	2.99***	0.29	3.99***	0.03	1.84***	0.03	2.84***	0.03	4.12***	0.03	3.74***	0.04
Intervention*Gender	0.89	1.9	0.29	1.4	0.15	0.14	0.1	0.13	0.08	0.13	0.02	0.14	0.19	0.18
Model R2 Gender	0.32		0.27		0.59		0.59		0.53		0.51		0.42
Intercept Childcare	2.78***	0.67	3.09***	0.44	3.91***	0.05	1.92***	0.05	2.84***	0.05	4.04***	0.05	3.67***	0.07
Intervention*Childcare	0.82	0.95	-0.12	0.61	-0.07	0.07	0.03	0.06	0.03	0.07	-0.03	0.07	-0.03	0.1
Model R2 Childcare	0.34		0.27		0.59		0.59		0.52		0.51		0.42
Intercept Eldercare	1.84***	0.39	2.96***	0.29	3.98***	0.03	1.85***	0.03	2.84***	0.03	4.12***	0.03	3.74***	0.04
Intervention*Eldercare	-0.59	2	0.61	0.65	-0.05	0.07	0.04	0.08	0.06	0.08	0.06	0.09	0.07	0.12
Model R2 Eldercare	0.37		0.27		0.6		0.59		0.52		0.5		0.42
Intercept Sandwiched	1.15***	1.2	2.78***	0.43	3.94***	0.05	1.93***	0.05	2.86***	0.06	4.09***	0.05	3.81***	0.08
Intervention*Sandwiched	-1.09	1.5	-0.21	0.53	-0.05	0.07	-0.02	0.07	0.06	0.08	0.05	0.07	0.12	0.1
Model R2 Sandwiched	0.35		0.27		0.59		0.61		0.53		0.5		0.44
	Use of Family Leave		Use of Sick Leave		Leave Supportive Supervisor Behaviors (LSSB)		Psychological		General		Job		Organizational
6-months							Distress		Health		Satisfaction		Commitment
Variable	b	SE	b	SE	B	SE	b	SE	B	SE	b	SE	b	SE
Intercept Gender	2.31***	0.41	3.03***	0.4	4.03***	0.03	1.82***	0.03	2.87***	0.03	4.10***	0.03	3.66***	0.04
Intervention*Gender	0.49	1.8	-0.9	1.7	0.11	0.14	0.13	0.13	0.2	0.11	0.04	0.14	.39*	0.18
Model R2 Gender	0.08		0.05		0.59		0.53		0.61		0.51		0.39
Intercept Childcare	4.07***	0.9	3.93***	0.79	4.02***	0.06	1.84***	0.05	2.84***	0.05	4.07***	0.06	3.67***	0.07
Intervention*Childcare	1.12	1.4	2.09*	1	0.08	0.07	0.03	0.07	.13*	0.06	.18*	0.08	0.19	0.1
Model R2 Childcare	0.14		0.09		0.58		0.53		0.62		0.53		0.4
Intercept Eldercare	2.34***	0.41	3.13***	0.4	4.02***	0.03	1.82***	0.03	2.87***	0.03	4.09***	0.03	3.67***	0.04
Intervention*Eldercare	1.09	1.5	-0.76	1.2	0.11	0.08	-0.1	0.08	0.06	0.07	0.08	0.09	0.08	0.12
Model R2 Eldercare	0.08		0.09		0.59		0.53		0.6		0.51		0.39
Intercept Sandwiched	3.53***	0.97	3.55***	0.79	4.02***	0.06	1.81***	0.05	2.83***	0.05	4.06***	0.07	3.72***	0.09
Intervention*Sandwiched	0.01	1.3	0.49	1.1	0.01	0.08	0	0.07	0.11	0.07	.21*	0.09	0.13	0.12
Model R2 Sandwiched	0.11		0.06		0.58		0.53		0.62		0.54		0.4

Note. All models controlled for baseline outcomes, age, childcare role, and education. All estimates represent unstandardized values.

^*p < .05.

^**p < .01.

^***p < .001.

Gender

Employee gender (0 = men, 1 = women) significantly moderated the intervention on employee reports of organizational affective commitment at Time 3 (b = .39, SE = .18, p = .034, pseudo ΔR² = .010, 95% CI [.03, .75], such that employees’ who identified as women had increased commitment 6-months post-baseline, partially supporting H5. See Figure 3a.

Extent of Responsibility for Managing Caregiving Demands

Extent of responsibility for managing childcare demands.

Prior to conducting moderation intervention effects, preliminary analyses were conducted to see if there were gender differences in child caregiving role (yes/no) and the extent of responsibility involvement. Results indicated that there were no gender differences in the proportion of employees who reported having children under the age of 19 living at home, 72 (42.60%) men, 226 (45.02%) women, χ² (1, 671) = .30, p = .58, However, women reported significantly more childcare responsibility involvement, M_women = 3.17, SD_women = 1.39, M_men = 2.75, SD_men = 1.30, t(307) = -2.28, p = .02. Given these differences, we used the continuous measure of extent of responsibility for care demands, rather than simply a dichotomous measure of having a dependent or not, in analyses.

Employee baseline perceptions of extent of childcare responsibility involvement significantly moderated the intervention on employee reports of job satisfaction at Time 3 (b = .18, SE = .08, p = .024, pseudo ΔR² = .032, 95% CI [.02, .33], employee reports of general health at Time 3 (b = .13, SE = .06, p = .028, pseudo ΔR² = .030, 95% CI [.01, .25], and employee reports of paid sick leave usage at Time 3 (b = 2.09, SE = 1.02, p = .041, pseudo ΔR² = .031, 95% CI [.09, 4.10]), such that employees’ who reported high responsibility for managing childcare demands at baseline had increased reports of job satisfaction and general health 6-months post-baseline compared to those with lower levels of responsibility. High childcare responsibility involvement at baseline did not moderate any of the other outcomes, partially supporting H6a. See Figures 3b, 3c, and 3e.

High responsibility for managing older adult care demands.

Preliminary analyses revealed that women employees were significantly more likely to report caring for an older adult (196, 38.6%) relative to men employees (45, 26.6%), χ² (1, 677) = 7.91, p = .005, However, among those in an adult care role, there were not significant gender differences in older adult care responsibility involvement, M_women = 2.86, SD_women = 1.29, M_men = 2.53, SD_men = 1.29, t(239) = -1.57, p = .12. Employee baseline perceptions of elder care responsibility involvement did not significantly moderate the intervention on any of the associated outcomes of general health, LSSB, psychological distress, organizational commitment, job satisfaction, or usage of leaves. Thus, Hypothesis H6b was not supported.

High responsibility for managing both childcare and older adult care demands (sandwiched care).

Preliminary analyses revealed that women employees were significantly more likely to report caring for both a child under the age of 19 living at home and an older adult (i.e., sandwiched care; 108, 21.4%) relative to men employees (23, 13.5%), χ² (1, 675) = 5.02, p = .03, However, among those in a sandwiched care role, there were not significant gender differences in responsibility involvement, M_women = 6.11, SD_women = 2.24, M_men = 5.84, SD_men = 2.04, t(129) = -0.53, p = .60.

Employee baseline perceptions of high responsibility involvement for managing both childcare and older adult care demands significantly moderated the intervention on employee reports of job satisfaction at Time 3 (b = .21, SE = .09, p = .024, pseudo ΔR² = .064, 95% CI [.03, .39], such that employees’ who reported higher responsibility at baseline had increased job satisfaction at 6-months post-baseline compared to those with lower responsibilities. Responsibility for managing both childcare and older adult care demands at baseline did not moderate any of the other outcomes. See Figure 3d. Hypothesis H6c was partially supported.

Post hoc analyses on mechanism: Supervisor role modeling

Although the intervention produced main effects on several study outcomes (e.g., LSSB, organizational commitment, job satisfaction, and leave use), mediation analyses clarified that these effects did not operate through the global composite employee perceptions of LSSB as initially theorized. To clarify the mechanism through which the intervention operated, we conducted post-hoc longitudinal mediation analyses on the sub-dimensions of supervisor supportive behaviors as other studies have done (Odle-Dusseau et al., 2016). The post hoc analyses noted in further detail in Supplement Appendix 6: Post-hoc Tests indicate that the intervention’s impact was transmitted specifically through increases in employee- reported perceptions of supervisor role modeling behaviors at Time 2, which in turn accounted for a meaningful portion of the intervention effect on both employee perceptions of LSSB at Time 3 (ab = .053, SE = .03, 95% confidence interval (CI) = [.007 to .109], αβ = .028) and employee job satisfaction at Time 3 (ab = .029, SE = .02, 95% confidence interval (CI) = [.001 to .085], αβ = .016), suggesting that this behavioral sub-dimension represents a particularly proximal JD-R resource. Indirect effect tests supported this sequential pathway that the intervention strengthened a specific behavioral resource of supervisor role modeling behaviors for leave that subsequently enhanced employee attitudes and perceptions of leave-supportive supervisory behaviors. Other LSSB dimensions did not demonstrate comparable mediation effects.

Discussion

In this study, we integrated a Job Demands-Resources (JDR) theoretical perspective with research on family supportive supervision, and work-family policies to develop and evaluate the effects of a training intervention targeting Leave Supportive Supervisor Behaviors (LSSB) on paid leave use, and work and health outcomes. Compared to employees in the control university (no trained supervisors), employees in the university with trained supervisors reported higher perceptions of LSSB (Time 3, 6 months post-baseline), greater family leave use (Time 2, 3-months post-baseline and 3), higher job satisfaction (Time 2), and stronger organizational commitment (Time 2). There were no significant main effects for psychological distress, general health or sick leave use.

Moderated effects indicate that the supervisor training was particularly beneficial for women (higher commitment, Time 3), employees with childcare demands (greater job satisfaction, better general health, sick leave use, Time 3), and sandwiched caregivers (higher job satisfaction, Time 3). Thus, training supervisors to support paid leave not only benefits traditional constituents with greater “need” (women, parents, sandwiched care) but, given main effects, enhances value across the entire workforce.

Supervisor Role Modeling of Paid Sick and Family Leave Use: Possible Mechanism

Post-hoc analyses suggest that the training intervention operated through the role modeling dimension of the LSSB. Training supervisors increased employees’ perceptions of supervisor role modeling behaviors at Time 2, which in turn predicted higher employee-reported LSSB and job satisfaction at Time 3. Trained supervisors may be more motivated to exhibit role modeling to use leave which influences employees’ perceptions of supervisor support for leaves as resources to manage work-family demands, which also benefits job satisfaction. This helps close the gap between formal leave policy availability and informal support as supervisor role modeling strengthens work-family supports as contextual resources. Supervisor role model behaviors may serve as proximal resources that reduce normative uncertainty about using leave activating motivational pathways consistent with JD-R theory. The intervention may have influenced outcomes by altering normative signals, reducing perceived interpersonal or career costs associated with leave, which are mechanisms embedded within supervisory role modeling. A caveat of these findings is that because the indirect effects were modest and direct effects were attenuated in mediation models, additional proximal psychological resources (e.g., perceived organizational support for taking and using leave, clarity about leave use policies, or reduced anticipated leave stigma) may also contribute to the intervention’s effects.

Theoretical Implications

Our study advances theory linking JDR theory with family-supportive supervision and paid leave research to advance integration of research on formal and informal work-family support in several ways. First, JD-R theory traditionally assumes that enhancing job resources (training to motivate supervisors to increase their work support) creates a supportive context for employees to manage job demands. This study shows that LSSB training increases employee awareness of leaves as formal work resources for managing work-family demands, providing empirical evidence to support theorizing on J-DR theory and work-family linkages (ten Brummelhuis & Bakker, 2012). Organizations that pair supervisor training with formal leave policies strengthen employees' perceptions of LSSB, with supervisor role modeling as a mechanism to enhance work and health outcomes and supervisor motivation to role model leave support. As intervention research grows, JD-R theory can guide theorizing on how interventions strengthen supervisors role modeling of organizational support in the work context as resources benefiting both work outcomes and nonwork outcomes.

Consistent with JD-R theory, role modeling may function as a social resource that enhances the usability and legitimacy of formal work-family policies (e.g. leave), activating motivational pathways that improve employee work perceptions. Studies suggest that employees may learn health and unhealthy work behaviors such as presenteeism and working when sick from supervisors (Dietz et al, 2020). Another study in the hospitality industry found that employees are more likely to work when sick when they distrust supervisors (Leal, Ferreira, & Carvalho, 2022). Role modeling use of leave policies may enhance employee perceptions of trust in their ability to access and use these policies without backlash and could enhance the overall firm health. When supervisors role model of healthy behaviors to use leave, through social learning, it provides a “behavioral blueprint” for employees to access and use leave.

Consistent with past research on family supportive supervisor behavior sub-dimensions (Odle-Dusseau et al., 2016), future research should investigate under what conditions the various LSSB dimensions are more important in their impact on work, health, and leave use outcomes and for whom and under what conditions. Future research also should incorporate more proximal mediators and finer-grained temporal measurement to better capture the unfolding resource activation process central to JD-R theory.

Our findings add to work-family research and theory theorizing by reinforcing the importance of proactively socializing supervisors to set norms to countervail the persistent decoupling between formal and informal work-family support. While supervisor support has been identified as important for policy take up in work-family studies, organizational-wide interventions that proactively train leaders to support leave (a time off policy) have been under-integrated into mainstream research on perceived organizational support (POS) for the work-family interface. Our study also suggests that future theory enhancing organizational support for paid leave and other work-family policies may have different meaning and value for different employee groups’ needs and identities. These findings on the varying meanings and dynamics might be strengthened into the design of future interventions. Studies might assess whether and why employees want to reciprocate for supportive environments providing resources of time off for personal life and family care needs.

Building on previous research showing that work-family-specific support is more likely to benefit employees’ work-family needs than general support (Kossek et al., 2011), we add to family supportive supervisor behavior research. Past workplace family-supportive supervisor supportive behavior interventions have integrated other health domains, such as sleep and safety, to benefit employees' job and personal well-being (e.g., Brossoit et al., 2023; Hammer et al., 2021). However, no measurement development or intervention studies have explicitly focused on enhancing supervisor support for specific work-family policies, such as paid family and sick leave. This is a critical theoretical and empirical contribution, given that leave is often underused, undervalued, and many employees are unaware of how to access it or are afraid to do so due to potential job-related negative consequences (Davison & Blackburn, 2023).

Practical and Policy Implications

This study has important policy implications for many stakeholder groups: employees, employers, supervisors/leaders, human resources experts in health, benefits, work-family, leadership development, recruitment, retention, and State and Federal government policymakers.

Implications for Employers

Organizations should proactively support the implementation of paid leave policies by integrating leave support into supervisor and leadership development. Our findings reveal the importance of interventions increasing supervisors’ awareness and skills to exhibit supportive behaviors – particularly to model taking care of their own health, for leave appears to significantly improve employee leave access and utilization. Training supervisors to support paid leave appears to close the persistent gap between paid leave availability and use and implementation effectiveness (Buys, Selander, & Sun, 2019; Geraldes, Chambel, & Carvalho, 2024) . For example, after passage of California’s paid leave policy, which requires employers to provide up to six weeks of paid leave to care for a new child or seriously ill family member, over half of eligible employees did not know about the policy and over a third who knew about the benefit did not use it due to fear of negative consequences (Appelbaum & Milkman, 2011).

Offering supervisor training enhances the value of leaves as a contextual resource for all employees, and especially those with higher care demands. This makes paid leave seen as more "usable" and personally relevant resource, strengthens employer-employee relationships, and improves job and health outcomes. Yet currently, the lack of workplace support for leave is a persistent social problem: namely the inability of caregivers and employees to take time off from work when they are sick or when their families need care. Our findings show employers gain productivity benefits from training supervisors to support paid leave: increased job satisfaction and commitment. Such findings can hopefully change the employer narrative on paid leave while showing the benefits of improving the health and well-being of all employees and their families.

Employers should consider instituting consistent, frequent, and timely leave communications such as during benefits enrollment periods and involving supervisors in departmental mandatory annual training. This approach shifts the paradigm of human resources as passive responder to an active promoter of leave. With widespread knowledge of leave policies at the organizational, supervisor, and employee levels, companies can retain valuable employees, including caregivers, and boost overall productivity by reducing stress and burnout as well as enhancing commitment, engagement (Wood et al., 2020), job satisfaction and retention (Waldfogel et al., 1999). While our training intervention focused on supervisors, work and health outcomes could be improved through implementing similar employee training on accessing supporting use, the importance and value of leaves (Dahl et al., 2014).

Embedding these practices into leadership development and talent fosters a work-life supportive context (Kossek et al., 2024), as leaders are equipped to recognize and address caregivers’ unique challenges enhancing job satisfaction and organizational commitment for individuals and across teams. Lastly, few employers involve supervisors in paid leave implementation or train them in supportive leave practices. Our study suggests both employers and employees would benefit if supervisors were more proactively trained with line managers- not only HR, actively supporting positive use and work and health outcomes.

Implications for Government

As the U.S. continues to rely heavily on voluntary employer provision of paid leave, government policies that encourage or mandate leave are needed. Our findings underscore the critical role supervisors play in effective implementation. Government initiatives could incentivize supervisor training and expand paid leave access, leading to a more resilient national workforce over time. Such measures could advance the health and equity benefits of leave and ensure minimal safety standards protecting those needing time off for family or personal illness.

Our study indicates that legally requiring paid leave policies is insufficient for equitable access and use. Governments should assist employers in fostering a work culture that improves knowledge and behaviors regarding leave (Cha & Grady, 2024). Evidence suggests that paid leave has significant health and well-being benefits for low-income workers and their families, reducing poverty and enhancing child and maternal health (Goodman et al., 2020).

Future Directions and Limitations

Our study shows that, even after controlling for FSSB, enhancing LSSB has specific employee benefits, opening a promising line of research on enhancing utilization of specific work-family policies through leader role modeling. Role modeling provides a theoretical bridge between often-siloed work-family support and leadership literatures (Kossek, Perrigino, Russo, Morandin, 2022). Future research might investigate whether leaders who role model using leave and other work-life policies are more transformational in their leadership style and trusted. Studies also might examine peer group role modeling of healthy work cultures that avoid presenteeism of working when sick and team level trust and productivity.

Future research needs to measure and theorize how to dually enhance formal and informal work-family support and examine how to better incorporate these linkages into interventions to reduce work-family conflict, enhance health, well-being and productivity. This study has illuminated the value of integrating JD-R theorizing and measures into work-family research intervention studies. The workplace has been conceptualized and studied more often as a support for work outcomes and a source of work-family conflict. Future studies can also foster a better understanding of how the organizational context can provide resources for family and personal life well-being through paid leave coupled with supervisor support, as well as other work-family supports (e.g., child and elder care, wellness policies, mindfulness, etc.). There are many possibilities to build upon.

Future research should also replicate and deepen analysis of our results regarding leave moderators, specifically, women and caregivers with greater responsibility for managing care while working who may be increasingly experiencing dual work and family role intensification. Scholars can build on our measure of primary caregiving responsibility, which captures primary role involvement of the worker-an improvement over many other measures. Future research should move beyond only using descriptive measures of family role occupancy (i.e., Are you a parent, elder care giver, married? Number of children?) uses in many studies, which may currently overstate involvement in managing care. Studies might also assess the severity of care demands (e.g., mental load) (Wayne et al., 2023), number of hours, and activities of daily living being managed to identify those who are most likely to benefit from interventions. This could entail needs assessments of the different types of care demands employees are managing and are on call for during the day (e.g., autistic child, adult with dementia) to enable improved design of customized interventions. Additional understudied employee groups (e.g., physical disabilities, LGBTQ+ employees, or fathers who act counter to masculinity norms as primary caregivers) for whom linking needs assessments to interventions also could be especially beneficial.

Every study has limitations, and this study is no exception. One important limitation is the use of one university in each of the two study conditions. Although this design is commonly used, some methodologists argue a two-organization study raises questions on the internal validity of study conclusions (as existing organizational differences at baseline can mask intervention effects post-intervention )and the possibility for Type I error control for the intervention effect model parameters in statistical models (Murray et al., 2004; Varnell et al., 2001). Regarding threats to internal validity, we minimized these effects by including baseline values of our outcome variables. We also included organizational controls of any pre-existing differences that would affect these variables at baseline, and thus effectively account for these organizational differences when evaluating changes in outcomes across conditions at follow-up. We also accounted for additional differences we found across organizations at baseline in our models, which are not directly related to our focal outcome measures (see the Methods section, where we discuss organizational control variables). Thus, we have strong confidence that any differences observed at follow-up are attributable to the training intervention and not to pre-existing differences across organizations at baseline.

Second, methodological research indicates that failing to control pre-existing organizational differences, most often by including random effects for organizations within condition, can result in inadequate Type I error control for the intervention effect parameter (Varnell et al., 2001). We agree with this conclusion but note that the generating and analytical models in this research do not incorporate baseline values for the outcome variables, unlike our models. This study demonstrates that when these organization-level intraclass correlations are not accounted for, Type I error inflation ensues. However, when the organization-level ICCs are very small, no important Type I error inflation is observed. Our analytical models account for baseline organizational differences, and as a result, the residual ICCs are very small once baseline outcome variables and other organizational differences are controlled. As a result, we conclude, based on the Varnell et al. (2001) article, that our results are unlikely to be a function of Type I error inflation. We encourage future research to include more than one organization in each intervention condition.

Another limitation is our reliance on employee self-report data which can increase measurement error and bias (Chan, 2009). However, perceptions of leave support are essential to understanding employees' psychological responses to resource acquisition, as it is critical to determine whether employees believe their supervisors’ behaviors have changed post-intervention. Further, longitudinal data requires handling missing data, and we have clearly laid out the implications of missing data and attrition in our methods. Our treatment organization did not allow payment to study participants which likely affected response rate over time.

We used a rigorous intent-to-treat (ITT) approach, meaning all employees in the organization assigned to the training condition were included in analyses, regardless of whether their direct supervisors completed the training. This method, consistent with organizational intervention research (e.g., Kelly et al., 2014), provides a conservative test of effectiveness and preserves the integrity of the randomized cluster field study design while reflecting real-world variability in exposure. Also, unlike most workplace wellness studies, while only examine effects on direct participants, we evaluated distal benefits for employees whose supervisors, rather than themselves, received the training. This approach allows for rigorous testing of supervisor-focused interventions aimed at improving organizational implementation and addressing underutilization of paid leave (e.g., Chen, 2016).

Finally, our measure of affective organizational commitment revealed significant differences due to employee attrition. Specifically, participants included in the analysis differed significantly in their responses compared with those who only completed the baseline survey (Goodman & Blum, 1996). Single-item measures, limit precision in estimating true variance. They do not allow separation of true-score variance from error variance needed to assess reliability. While such error reduces statistical power to detect intervention effects greater error variance typically obscures true effects rather than producing false positives. Thus, when significant and theoretically consistent effects emerge despite reduced power, this outcome strengthens confidence in the findings and suggests that concerns about measurement error may be overstated. Nevertheless, because single-item measures limit precision in estimating true variance, we encourage future research to use the full 8-item affective commitment scale to improve power and attrition estimates (Allen & Meyer, 1990). Items 2, 6, and 8 have shown the highest inter-item correlations and may be used when survey length or fatigue are concerns. Yet one-item measures can be just as reliable and valid as multi-item scales (Matthews et al., 2022).

Conclusion

Kehoe and Han (2020) note that corporate-level HR policies often diverge from employees’ actual experiences, with substantial variation across organizations. This study shows that supervisors, typically under-emphasized in HR’s administration of leave, play a pivotal role in implementing work-family policies. Employees’ perceptions of supervisor support for leave are critical for realizing their benefits. Yet rigorous research on how to strengthen this support remains limited, even though most countries and employers now have paid sick and family leave policies (Glynn, 2023). Even if federal mandates for paid leave are enacted in the U.S., disparities in employee access and supervisor implementation are likely to persist. Our findings show that training supervisors to support paid leave is essential for improving employee health and work outcomes. Integrating informal supervisory support with formal HR policies enhances policy effectiveness. Replicating this work across industries and countries would further advance understanding and application of these issues, advancing work-life equality, well-being and productivity across society.

Appendix 1

A. HR Expert Semi-Structured Interview Protocol and Sample Responses

Semi Structured Interview Protocol

Probes were used to follow up answers

1. Please briefly tell me about your company’s policies on family care and sick leave policies

2. What do you see as managers’ role in educating employees about the policies and implementing these policies?

3. How do they inform employees about the policies?

4. Do managers get any training on leave use?

5. What do you see as the benefits of offering employees family care and sick leave?

6. What do you see as challenges of offering employees family care and sick leave?

B. Sample Background Quotes and Summarized Examples from Expert Interviews

Interviewee ID	Background	Leave supportive Supervisor behaviors	Comments on why Supervisors & Managers should care about leaves or need socialization/training issues	Misc other quotes
001	CEO HR Consulting Firm/ Former VP and Director of 3 Fortune 500 companies	“I do think as we evolve to be greater workplaces that want to provide more empathy and support, we should train managers [to communicate to employees], “I’m here to help shepherd you through this process.” “The manager should help the employee navigate getting what they need.” I think it’s educating the leader to make it safe for the employee to have a conversation that says, “I need an accommodation.”	“ “I don’t think that many executive non-HR personnel have really quantified the cost of replacing people.”	“The long game is to support people, so they’ll want to work there for the long haul.” It’s not the kind of policy that you can [implement] based on performance.”
002	HR Director of Benefits, Large Public University A	“In our [central HR] roles we promote all types of wellness, so taking leave, for whatever reason, for self or family, is very important and is part of overall health. ” We want managers to understand that, to appreciate that, and to encourage it.” “I would want a manager to know that when they're hiring a new employee, one, that there's a leave policy and if they don’t have it memorized, fine, … at least to be able to direct the employee to know, hey, you have sick time available, and you have vacation time available, or whatever the leaves are at their respective employer, and direct them there, hand it to them if that seems appropriate. ….I would expect and we need managers to know, that there are leave policies in place, and employees are entitled to that policy. They need to know about it, and they need to be able to use it and then how to use it. I think part of a manager's responsibility is to ensure that their employees know how to ask for time off and have that conversation of when XXX needs a day off, I should know, my manager, what he or she expects from me, as far as when I should ask for that.”	“For the most part, employees should be trusted and are trying to do the right thing.” “ Taking leave, for whatever reason, for self or family, is very important and is part of overall health….We want managers to understand that, to appreciate that, and to encourage it…. “We have some work to do on the HR side about communicating how healthy it is to take time away and how managers need to support taking time away.” “It benefits everybody.”	“It’s become part of our culture in HR that people should take time off when they need it for whatever reason they need it for. They should be left to do that without guilt, and nobody should be calling them when they’re off.” (additional behaviors noted: prevent stigmatization by supporting and respecting leave time off) We are there to support each other when life happens, at work or outside of work.” “Part of that deal is helping do things while people are out.” “You don’t necessarily have to have a destination to take time off of work.”
003	Manager of workers’ compensation, long term disability, FMLA, and parental leave programs, Large Public University B	We want supervisors and employees to have a good relationship. It’s okay to show human compassion and to ask, “How are you doing?” You are not asking about work. You can ask them, “How’s your recovery going?” or “Are you still planning to return to work on Monday?” Those things are okay to ask.”	On why Supervisors Need Training: “If you have never need to use it, you don’t know about it.” “We used to offer, I would say, a significant amount of training.. a class once a month. It was a full-day-long class about what to do if your employee was hurt. What to do if your employee is sick…. Due to low enrollment, we had canceled it maybe five years ago or whatever. Somewhat in place of it, we have implemented a new supervisor …essentials class…I would actually go in, and I would speak for half hour to 45 minutes trying to say, "Hey, heads up. This is some of the stuff that we offer. If you have questions, please ask us." However, due to employment law, we don’t necessarily want our managers and supervisors giving a ton of advice and giving the wrong advice. My message is, "Hey, I want you to be informed of this, but I don’t want you to be necessarily advising of this. If you want help out, we are all for that. If you don’t know the answer, I need you to really reach out to us. It’s totally okay to just refer the employee to us in the first place."	Regarding barriers to using family care and sick leave: “#1 [barrier] is education, [meaning] what the options are, or that there might not be an option for what they need.” Told Story about leave abuse: An employee applied for FMLA and was approved … They would be unavailable to work the second half of the week (i.e., Wednesday-Friday) because they needed to care for a family member. This employee’s manager took that Friday off as well. The manager ran into the employee at … vacation spot. Come Monday, disciplinary action was taken against the employee. Later in interview told another story about a worker… who said they were feeling better but not great and asked their doctor for a little more time off. Someone found out that they used that time off to finish… (outside remodeling) …at their home. “These plans are in place to help employees, and when we’re spending our time disciplining employees who are abusing it, it’s taking time from away from employees that really need it and at the same time it’s putting a negative stigma on these plans.”
004	HR Director at Manufacturing Firm (Additional interviews were also conducted with the recruitment manager & another HR manager).	“When I have people who-- I would have my supervisors call me and say, "XXX is out for a sprained ankle or something," and they are like, "God, I can’t believe it. She was on a ski trip, and she sprained her ankle and she can’t come in." It’s like, 'You know what? It doesn’t matter but you still have to do it." (support leave) Sometimes it’s calming the nerves of the supervisor and not getting the rest of the team excited about somebody being out. You have to remember that life happens to everybody. It’s their turn to have something maybe bad happen and we have to pick up the pieces around that. We don’t have-- These are things that can be out of our control, but let’s not make it worse.”	There's nothing like an employee that’s distracted by not having childcare… If you can’t take care of your family, then you can’t take care of your job either. To me, when people at XXX are saying, "Oh my gosh. Thank you for giving me time off to take care of this or that," they just seem more dedicated and are more invested in the organization overall” Interviewer: What is the manager's role in educating and communicating these policies to employees? HR Director: I don’t think we really train our managers to do that. Interviewer: That was the next question. [laughs] HR Director: We fly by the seat of our pants. We might talk to a peer to find out what we should do. I've done that before like, "I got this situation with an employee, what should I do?" I know who to go to, but I also have people who don’t have good networks at XX either they're new or inexperienced. The thing I hate the most is when I hear somebody say, "I didn’t know what to do, so I did nothing. “If we can teach people what their role is-- The other thing is, they'll also say, "Well, I didn’t know because it’s like a personal thing. Should I get involved or not?" or "I don’t want to get involved." I think there is some responsibility on the supervisors to be involved, or the manager. Even if that’s to say, "Go contact HR," or "Go read the handbook," or "Look at our policy."	- We don’t apply our attendance policy to the way it should be applied. The way it’s written. Interviewer: What do you mean? HR Director:: I mean that we don’t-- They're supposed to take certain days just for sick time, and then other days for-- You have to book one week of vacation. One solid week of vacation, and then you can take it intermittently. You're not supposed to take call off time before holiday and after holiday. … We can’t have 10 people calling in on the day before the holiday.
005	Director of Compensation & Benefits at Public University C	I just think that the communication part of it is how that employee is going to feel like they're not ostracized, and that they're connected still to the department. If the manager is reaching out periodically to say, "Hey, just checking on you to see how you're doing," without making that person feel like they have to give you a lot of details with regard to what they're actually going through in treatment. ..it’s a simple contact. It could be a text message, it could be an email, or phone call to say, "Thinking about you. Just want you to know that we're supporting you. Keep in touch and let me know when you're ready to come back to work..	We do have some opportunities where we do need to get out and do a lot more manager training, because we do have a lot of new managers. What we would like for them to do, and what we expect for them to do as stakeholders in the process, is to basically make sure that if they hear something, because of course, if an employee doesn’t know that they need to call the leaves center, and if they hear anything with regard to something that may qualify for any of these benefits, to refer them to us, number one, but to also support them while they're out on their leave event.; We really need managers to help us with regard to making sure employees are being communicative with regard to their expected return to work dates, making sure that they are submitting their release to return to work. It’s not a matter of them approving or denying, we have completely removed that burden from the managers, that’s why my team is here, but they really need to engage in the process and make sure that they're helping with those loose ends of tying up the return-to-work requirements	“I can tell you that with me and my team, and it’s unfortunate, because a lot of my time is spent on things that are getting escalated, because a manager has done everything that they can to try to get an employee to follow the rules and do what they're supposed to do, but then it gets escalated. That’s where we spend our time, or if it’s on documents that have been forged or falsified, so then we have to investigate that….” The only best practice that I can say, which is what we recommend to managers, is to make sure that they are holding employees accountable to the approvals that we're granting them. For instance, intermittently, there's anybody you talk to that that manages leave, they will tell you that that’s where the nightmare lies. Intermittent leaves are very hard to manage. I'm sorry.”