Research-Supported Intervention and Discretion Among Frontline Workers Implementing Home Visitation Services

Abstract

Objective

We examine how frontline workers and supervisors delivering a research supported intervention (RSI) to reduce child neglect negotiated system-related challenges, the pragmatics of RSI implementation, and their professional identities and relationships with clients.

Methods

We conducted semi-structured interviews, small group discussions, and focus groups with frontline workers and supervisors in one large county over two time periods. We used iterative coding to analyze qualitative data.

Results

Frontline workers navigated several aspects of RSI implementation and sustainment: (1) contract requirements and information dissemination, (2) fidelity, (3) competing demands and crises, (4) structure versus creativity, and (5) relationships with clients.

Conclusions

Workers dynamically negotiated multiple system- and provider-level (or outer- and inner-contextual) demands influencing RSI provision for clients with complex service needs. Results affirm the need to attend to the unintended consequences of implementing new contract, reimbursement, and other system organizational processes and to address the “committed work” supporting RSI delivery.

The provision of research-supported interventions (RSIs) within human service sectors does not reflect a straightforward unidirectional movement of evidence to practice; rather, it is more fittingly characterized as an ongoing process of knowledge exchange shaped by the context-contingent actions of frontline workers tasked with implementation (Aarons and Palinkas 2007; Barwick et al., 2005; Kimber, Barwick, & Fearing, 2012; Palinkas et al., 2009). The real world use of RSIs does not occur within controlled laboratory settings, but within complex service delivery systems in which multilayered bureaucracies may lead implementation efforts, and contracts between public entities and community-based organizations (CBOs) define the parameters of RSI provision (Ferlie & Shortell, 2001). Ongoing implementation within these systems is further complicated by the shifting and uncertain nature of financial resources to support RSI delivery (Aarons and Palinkas 2007, Fixsen et al. 2005, Greenhalgh et al. 2004).

The delivery of home visitation programs and other frontline human services represents a paradoxical form of labor in that the performance of paid work “tend[s] to slip easily into the discourse of kinship, caring, [and] labors of love” (Hopper 2006 p. 221). Frontline workers, such as caseworkers and mi-level human service professionals (Lamphere, 2005) who are enlisted to deliver RSIs to clients with complex needs, commonly find themselves engaging in “committed work”, referring to “labor performed under the auspices of a service contract or salaried job that goes well beyond the call of duty” (Hopper 2006 p. 221). These labors can take varied forms, from the provision of unbilled services and unpaid overtime to often exhausting emotional involvement in clients’ lives (Aarons, Fettes, Flores, & Sommerfeld, 2009) and myriad unquantifiable wasy of “going the extra mile” (Lamphere, 2005, p. 15). In this article, we explore the relationship between committed work and RSI delivery.

The alignment of RSIs with the existing values and ethics of care of frontline workers helps facilitate both their implementation and sustainment (Aarons and Palinkas 2007, Palinkas et al. 2009). Yet, the burden of this alignment may fall mostly to these committed workers and their direct supervisors who are responsible for ensuring that the job activities of home visitors adhere to RSI contracts typically negotiated between upper-level CBO administrators and public officials. In this article, we analyze the nuanced ways in which frontline workers and their supervisors interpret, experience, and reconcile the demands of RSI implementation with their own ideas of professional and ethical service delivery in hierarchically organized human service milieus. Moreover, we elucidate factors intrinsic to the successful implementation and sustainment of RSIs with potential to improve both the quality of human services and client outcomes (Horwitz et al. 2010, Mildon and Shlonsky 2011).

Approaches to RSI Frontline Service Work

Conceptual models of RSI implementation emphasize both the inner organizational and outer service system contexts of service delivery (Aarons, Hurlburt, & Horwitz, 2011; Willging, Sommerfeld, Aarons, & Waitzkin, 2014). The inner context encompasses the attitudes, behaviors, and actions of frontline providers and their supervisors. It includes commitment to the RSI, fidelity or the extent to which the RSI is implemented as intended, job satisfaction and turnover, and factors specific to CBOs, such as structure, climate, culture, and leadership. In contrast, the outer context directs attention to factors at the system level and comprises both interorganizational relationships between funders and CBOs charged with RSI implementation and the broader service delivery environment. This environment is itself affected by requirements written into policy, legislation, and contracts (Aarons, Sommerfeld, & Willging, 2011). Both inner- and outer-context factors shape the everyday tasks, interactions, and subjective experiences of frontline workers (Willging et al. 2014). Qualitative analysis makes it possible to identify and assess the respective implications of these factors for large-scale implementation of RSIs in human service systems.

The literature on frontline service work in human service systems has often focused on the ways in which workers use discretion, or on-the-ground decision making, to negotiate inner- and outer-context factors in their everyday work (Brodwin 2012, Durose 2009, McCracken and Marsh 2008, Prior and Barnes 2011). Scholars have emphasized that workers use reflection and critical thinking rather than rote practice when delivering services (McCracken and Marsh 2008). Workers have thus been characterized as “situated agents” who “act entrepreneurially” that is, independently and in their own interests (Durose 2009, p. 36), to make judgments about how to “reconcile” outer-context factors, such as policy prescriptions and contract requirements, with “their own personal, value-laden assessment of context-specific needs and problems and their evaluation of the consequences of different possible responses to those needs and problems” (Prior and Barnes 2011, p. 267). These perspectives often draw on Michael Lipsky’s (1980) concept of the “street-level bureaucrat,” the quintessential frontline worker who copes with an inner context marked by increased workload and scarce resources by deciding which parts of his/her job to enact, creating de facto or informal policy in the process. Recent scholars have refined Lipsky’s concept to show that street-level bureaucracy may not distort policy prescriptions, but rather function to mediate their effects on the inner context in practical ways (Prior and Barnes 2011)

The use of judgment and discretion by workers is particularly likely in frontline service work, where “policies rub up against strong values or interests” (Ellis 2011 p. 228). Deployment of discretion at the frontlines of human service systems can be expected in environments marked by the increased application of managerial techniques drawn from business environments, such as more stringent reimbursement procedures and protocols for reporting and performance monitoring. These techniques often place greater demands on employees, potentially limiting staff capacity and imposing financial constraints on frontline workers (Brodwin 2012, Ellis 2011, Willging et al. 2014). The relationship between outer- and inner-contexts can influence how direct service providers feel about their work lives and professional identities. For example, frontline workers may perceive a disproportionate emphasis on “quantity” over “quality” of work by organizational leadership or contract requirements can contributing to job dissatisfaction and feelings of injustice (Robins 2001), as well as concerns about the erosion of professionalism (Ellis 2011). In extreme cases, workers have struggled with a sense of futility as they contend with organizational demands that divert their time away from client care (Brodwin 2012).

Yet, even under demanding conditions, frontline workers may report greater satisfaction when they experience good working relationships and a sense of value in their jobs (Washington et al. 2009). They may also find creative ways to reconcile job requirements with their own ideas of the “proper way to live and to bear responsibility to others” (Brodwin 2012 p. 19). The moral and ethical positions of frontline workers are rarely considered in studies pertaining to RSI practice. We seek to elucidate how ethics of work and care are embedded in and negotiated within a nexus of inner- and outer-context concerns to illuminate the conditions and outcomes of RSI implementation for frontline workers.

Study context

The setting for this research is the child welfare system in a densely-populated county in the United States, which we will call “SafeCare County.” Here, a unique partnership was successfully forged between the local child welfare agency (termed County), nonprofit CBOs contracted to deliver home visitation services, and a private foundation, which we refer to as “Children Come First,” that provided some of the funding to support the initial implementation of the RSI, SafeCare (Aarons et al. 2014, Hurlburt et al. 2014). This manualized, highly structured RSI is delivered in the home setting to improve parenting skills for caregivers identified as at-risk or who have been reported for child maltreatment. Studied in over 60 scientific publications, SafeCare primarily targets families with children ages zero to five (Chaffin et al. 2012a, Gershater-Molko et al. 2003, Lutzker 1998). This increasingly popular RSI is currently implemented in several child welfare service systems nationwide (Gershater-Molko et al. 2003).

SafeCare delivery requires personnel in two primary roles: home visitors and coaches. Home visitors are specially trained in the RSI to deliver the in-home parenting curriculum, which is comprised of three modules: Health, Home Safety, and Parent-Child Interaction/Parent-Infant Interaction. Modules may be administered in any order, according to the primary needs of the family. Each module consists of six ordinal sessions that include role-playing, hands-on demonstrations, and assigned homework. All modules begin with baseline assessment sessions, followed by intervention (in-home training) and finally follow-up assessment to monitor progress toward the goals of the module.

SafeCare includes an overarching focus on structured problem solving and communication skills throughout the three modules. SafeCare coaches provide implementation guidance to the home visitors. They also accompany home visitors on monthly visitations to ensure fidelity to the RSI, using structured checklists to document the extent to which the home visitors maintain fidelity to the core elements of the specific SafeCare modules. As part of a larger research study on SafeCare implementation and sustainment (Aarons et al. 2014, Hurlburt et al. 2014), a second module-specific fidelity checklist, which parallels the one administered by coaches, is completed by clients at the end of each session. The checklist elicits data on activities occurring during the home visit that day. Coaching is intentionally distinct from general supervision provided to CBO employees, being advisory rather than supervisory. In SafeCare County, supervisors employed by CBOs manage the work of the home visitors, but are not required to be certified in the RSI. The National SafeCare Training and Research Center (NSTRC) sets training and certification standards. In SafeCare County, most home visitors and coaches have Bachelors’ Degrees and may or may not have prior experience working as case managers. Individuals may find themselves shifting between the roles of home visitor and coach if they possess the necessary credentials; therefore, we include persons occupying both types of positions in the sample for this study.

SafeCare services are facilitated by contracts between the County and four CBOs, each with their own organizational culture and climate and administrative practices. SafeCare was first implemented in 2008, in the middle of a five-year contract period that began in 2005. A second contracting period was initiated in 2010, after funding from Children Come First to support the initial training for and roll out of SafeCare had ended. SafeCare implementation is facilitated by the use of an interagency model that convenes service providers from the CBOs into teams to train them in effective RSI delivery within the service system (Hurlburt et al., 2014). A single individual – the SafeCare Director – helps coordinate program coaching and training. Individual home visitors are employed by one of the four CBOs contracted by the County. However, expertise of the SafeCare Director is shared across all four CBOs. Similarly, coaches at one CBO may be enlisted to work with home visitors based in other CBOs.

Research Design

Data Collection

As part of a longitudinal, mixed-method, multisite investigation of SafeCare implementation and sustainment, we collected qualitative data through 10 individual interviews, 4 small group interviews (i.e., < 5 participants), and 8 focus groups (i.e., 5 or more participants) in SafeCare County. Interviews and focus groups were conducted and analyzed by four anthropologists (including the first, second and fourth author) who were unknown to the participants and were not part of RSI implementation and sustainment efforts within the study system. Data collection focused on participant involvement in SafeCare at Time 1 (T1), the initial contracting period (2005–2010), and Time 2 (T2), the second contracting period (2010–2015). Starting in 2008 through 2012, we collected data each year across both time periods. At T1, the collaboration between the research team, public officials, and CBO management was a central part of the system-wide implementation. This collaboration facilitated access to home visitors and supervisors who met the inclusion criteria (employed by the CBOs and involved in SafeCare delivery). The home visitors and supervisors were invited to take part in this study by the study coordinator (third author) and did so voluntarily. At T2, the collaboration had been ongoing for several years, and the same approach was used to recruit home visitors and supervisors. No one declined for either time period. The home visitors and supervisors received a US$25 incentive for each data collection event.

We sampled key CBO supervisors (n=4 in T1 and 8 in T2), SafeCare coaches (n=2 in T1 and 4 in T2), and home visitors (n= 12 in T1 and 36 in T2). As illustrated in Table 1, the participants were overwhelmingly women, which is not atypical for the helping professions. Most were Hispanic, reflecting the demographics of workers in the larger contracting system and the service population of SafeCare County. The average age of participants fell in the mid-30s for T1 and T2, but the span was vast, especially for home visitors who ranged in age from 25 to 61, and who held their positions for an average of 3.5–4 years. All supervisors held master’s degrees, and most home visitors held bachelor’s degrees in social work, with others in psychology and related human service disciplines. As such, we should note that the majority of participants had training related to the ethical principles governing their respective discipline. For social workers, key values embedded in these principles include the commitment to service (helping people in need and addressing social problems) and competence (striving to enhance and apply professional knowledge and skills), challenging social injustice, emphasizing the dignity and worth of clients, recognizing the central importance of human relationships, and behaving in a trustworthy manner (National Association of Social Workers, 2008). For psychologists, key values center on beneficence and nonmalefiecence, trust, integrity, responsibility, and fairness in service provision (American Psychological Association, 2010).

Table 1.

Demography Information of Participants.

	Supervisors Coaches Home Visitors
Time 1
Female	100%	50%	92%
Age (years)	37.3	35	36.9
Race/ethnicity
Black	—	50%	17%
Hispanic	50%	—	75%
Non-Hispanic White	50%	50%	8%
Education
High school diploma	—	—	8%
Bachelor’s degree	—	50%	58%
Master's degree	100%	50%	33%
Discipline/training
Social work	50%	—	42%
Psychology	—	50%	25%
Child development	50%	—	25%
Other	—	50%	8%
Job tenure (years)	7.6	3.7	4.0
n	4	2	12
Time 2
Female	88%	100%	91%
Age (years)	41.4	33.3	33.6
Race/ethnicity
Black	—	25%	11%
Hispanic	25%	75%	80%
Non-Hispanic White	75%	—	3%
Another race/ethnicity	—	—	6%
Education
High school diploma	—	—	6%
Bachelor’s degree	—	75%	74%
Master’s degree	100%	25%	20%
Discipline/training
Social work	75%	50%	69%
Psychology	25%	25%	14%
Child development	—	25%	11%
Other	—	—	6%
Job tenure (years)	3.7	3.0	3.6
n	8	4	35a

^an % 35 for T2 home visitor group, with missing data on one participant; groups are not mutually exclusive (i.e., coaches overlap with home visitor group).

We used interviews and focus groups for efficiency and to ensure adequate representation of the various stakeholders. The semi-structured interview and focus group guides for T1 and T2 consisted of open-ended questions tailored to each stakeholder group. Supervisors and home visitors were purposefully insulated from one another during these discussions. Additionally, while some home visitors also worked part as coaches, none of these individuals participated in focus groups with staff whom they had directly coached. Each interview and focus group lasted approximately 60–90 minutes. We conducted the interviews and focus groups at CBO offices or at locations where team meetings were held. Questions largely centered on implementation challenges and inner-context issues, including worker knowledge, acceptance, and experience with SafeCare. However, the questions also elicited information on how participants identified and understood factors associated with the outer context that shaped their work and potentially affect RSI implementation and sustainment.

Data Analysis

All interviews and focus groups were digitally recorded, professionally transcribed, and checked for accuracy by at least one of the authors. We employed an iterative process to review this textual data while utilizing NVivo 10 qualitative data analysis software (QSR International 2012). Data analysis proceeded first by engaging in open and focused coding to locate themes and issues within the transcripts (Corbin and Strauss 2008). The transcripts were independently coded by two anthropologists (second and fourth authors) to condense the data into analyzable units. First segments of text ranging from a phrase to several paragraphs were assigned codes based a priori on the particular topic areas and questions that made up the interview and focus group guides. Sample codes included “collaboration,” “leadership,” and “contract.” During our review of the transcripts, open coding was then used to identify and define new codes to capture information on emergent themes (e.g., “ethics of care,” “client crisis,” and “acts of discretion”). Next we engaged in additional focused coding to determine which of these themes surfaced frequently and which represented unusual or particular concerns to the participants. The data were first coded by type of worker, and the data sets were compared.

In the staged approach to analysis, the two coders analyzed each transcript and created detailed memos that both described and connected codes to each theme and issue. Their work was then reviewed by the remaining authors, four of whom had studied the implementation and sustainment of the RSI for many years (third, fifth, seventh, and eighth authors) and one of whom was a public official who supported its integration within the public sector (sixth author). By constantly comparing and contrasting codes with one another (Corbin and Strauss 2008, Glaser and Strauss 1967), we grouped together those with similar content or meaning into broad themes linked to segments of text within our NVivo 10 database. A series of feedback sessions with participants later provided the opportunity to check the themes described in this article. While these sessions occurred more than 2 years after the T2 collection of data included in this analysis, participants recognized the themes described in this article and attested to their continued salience in understanding implementation issues surround the home visitation RSI within the more recent past.

Results

The final set of codes were constructed through a consensus of all the authors and condensed into a list of five major inter-related themes focused on how frontline workers navigate key aspects of RSI implementation. These themes were divided into two categories: (1) negotiations with the outer context and (2) negotiations within the inner context. The first category includes (a) contract requirements and information dissemination and (b) fidelity to the RSI. The second category includes (a) competing demands and crises; (b) structure versus creativity, and (c) relationships with clients. We describe each finding in detail below, and include quotations to illustrate significance. We edited quotations slightly to enhance readability.

Negotiations Within the Outer Context

Contract Requirements and Information Dissemination

Development and acceptance of contractual arrangements are crucial elements of the outer context that affect the way frontline workers and their supervisors perform their jobs. Contracts shaped day-to-day operations within the inner context and set outer-context administrative and reimbursement expectations for SafeCare delivery. Although these arrangement fell under the purview of upper-level administrators at the County and CBOs, home visitors and supervisors nonetheless reported that the terms of these contracts profoundly affected their experiences and that they remained a source of unease across the two time periods, T1 and T2.

Contracts at the start of T1 specified that CBOs received specific rates of reimbursement for each type of activity performed (e.g., in-home visits, client phone calls, and documentation and record-keeping). During T1, SafeCare was incorporated into existing contracts between the County and CBOs, the latter of which were enlisted to provide a variety of child welfare services, including but not limited to home visitation. The contracts included expectations that added to the workload of the home visitors who were also asked to engage in activities of the larger research study, such as participating in web-based surveys, interviews or focus groups, and facilitating completion of client-reported fidelity surveys. This last task slightly extended the length of home visits.

Repeated changes to contracts during the early implementation period were characterized as stressful, unclear, and overwhelming by both home visitors and their CBO supervisors. Supervisors were charged with ensuring that home visitors followed contract specifications correctly and struggled with multiple changes to protocols and processes related to formal documentation, communication, and the research study, combined with ongoing training, meeting, and fidelity checks. They contended that these shifting requirements forced them to place unreasonable demands upon the home visitors. They also generated additional work for supervisors, as they had to sift through what were perceived as conflicting “blast of emails” from upper-level administrators and the research team for final decisions to communicate to the home visitors. At the start of T2, one supervisor observed,

The last year or two, it’s really mellowed out. The amount of changes has really slowed down, but in the beginning, it was like every other month we had different outcomes, different expectations, different paperwork, and different ways of doing things. We’d get direction and then we’d have to get our staff to do it and just when they buy into it, they’d say, “Not really. We need to do it another way.”

For this reason, some supervisors described having home visitors continue doing paperwork that was no longer necessary to avoid “flip-flopping” if it was again required, even though this strategy created unnecessary work for home visitors. One supervisor explained, “It’s better to have more work that they understand and to have consistency than to have this up-down, up-down of trying to go with the decisions that were implemented.”

At the same time, a complex hierarchy of outer-context authority emerged from the considerable collaboration among multiple levels of stakeholders, including County officials (government), Children Come First (funder), and diverse CBO leadership (provider). This hierarchy sometimes fostered confusion in the inner context. In T1, home visitors and supervisors were unsure about the specific roles and responsibilities of key players in SafeCare, which led to trepidations about whose information related to the RSI and contract obligations was the most reliable. One home visitor said, “It’s kind of weird because there’s some things you can talk about [with your supervisor], but then it’s something that’s SafeCare and they say, ‘Well you’ve got to talk to [the SafeCare Director]’ and you talk to [the SafeCare Director] and they say, ‘You’ve got to talk to [your team leader].’ So you’re kind of guessing.”

In T2, these concerns were largely resolved. However major contract modifications occurred, which resulted in a change in how activities pertinent to SafeCare delivery were funded. The T2 contracts were negotiated by upper-level administrators at the County and the CBOs with the assistance of an outside consultant. Instead of the previous model of hourly reimbursement according to type of task, T2 contracts stipulated that CBOs be paid a significantly higher flat rate, but for “face-to-face” or in-home services only. The CBO leadership had advocated for this change. Along with County officials, they envisioned a equitable and simplified system under which home visitors received the same compensation but no longer have to itemize the minutiae of their work day. However, even after the CBOs had settled into the new contracts with the County, many home visitors reported experiencing “a lot of stress” about these modified requirements, particularly an increased pressure to log billable “face-to-face” hours. Several reported that they had to “chase clients” for billable time, and then rushed home visits so they might have time later in the day to complete required paperwork. As one home visitor described: “I feel sometimes that I’m in a hurry in a home visit and I need to get this client to put out the [fidelity] survey really fast because I need to leave. I need to go to the office and do paperwork.” A second home visitor explained, “We are only able to bill right now for face to face and phone calls with the social worker; before you could bill for the paperwork and other things.”

Despite the increased payment rate for in-home services, which was intended to better compensate CBOs, those on the frontlines lamented that half of the work day was devoted to unbillable activities, such as communicating with supervisors and County-employed child welfare social workers, and preparing materials for home visits. As a result, billing was considered to be “the number one stress factor for staff,” as a CBO supervisor observed. One home visitor explained how s/he spent less time preparing for visits because s/he could not bill that time: “If we just had the opportunity to prepare ahead of time, I would feel more confident and ready to go and provide the services, as opposed to feeling, sometimes, [that] we just didn’t have the proper time to prepare in advance.” Other participants echoed this frustration during T2. For instance, a second home visitor said, “I think it should be more about quality than quantity.” A third commented, “I would be a better case manager if I didn’t have to bill. When you’re in the home, you just want to make sure that you spend enough time there so you could go back to the office and have enough billing instead of talking to the client and addressing the other issues they’re having.” Home visitors critized what they perceived as demands to perform their jobs to achieve specific billable hours (“quantity”) at the expense of the clients (“quality”). In this way, many home visitors suggested that their understanding of contract requirements conflicted with their sense of how to best serve clients, created stress in the workplace, and compromised their ability to practice in accordance with the SafeCare model.

To deal with the increasing pressure to make and report billable, some home visitors reportedly performed extra work, such as tracking down client cell phone numbers to “make contact”. Home visitors also turned to supervisors to help them figure out how to bill their diverse activities in compliance with the contract. However, in keeping with their own sense of both personal and professional ethics of care, the majority of home visitors reportedly made decisions to simply work harder and for longer hours to adhere to contracts, while they simultaneously sought to address what they described as the “unbillable” needs of their clients.

Fidelity to the RSI

The highly structured elements of a RSI are stipulated by the intervention developers and require a high level of fidelity from frontline workers. Fidelity to the RSI is thus a second element of the outer context that shapes the experiences of workers and supervisors. Home visitors, coaches, and supervisors frequently voiced concerns about fidelity to SafeCare during T1 and to a lesser degree in T2. Although home visitors appreciated the support they received from fellow home visitors, coaches, CBO supervisors, and the SafeCare Director, many continued to feel that they sometimes received inconsistent guidance regarding how to deliver SafeCare with fidelity from these different sources. One home visitor in T1 complained, “When [the coach] comes, you ask a question. ‘Oh no this isn’t the way that you do it.’ But then we have [the SafeCare Director] another day and she says this is the way that you do it. And then you have someone else, they are not on the same page.” Similarly, a supervisor in T2 commented on different supervision practices across agencies, “Supervisors have different levels of involvement and sometimes it’s confusing I think. Sometimes there are questions that come up where they’re [CBO supervisors] unfamiliar with SafeCare and the model.” However, as supervisors became better acquainted with the actual practice of SafeCare, they tended to become more confident in their own ability to answer home visitors’ questions consistently.

Even though quantitative data collected for our larger study demonstrated high degrees of SafeCare fidelity across the CBOs (data available upon request), some confusion about the precise meaning of fidelity was evident among the participants in the study. Many home visitors and supervisors had come to understand the RSI as requiring far less discretion on their part to implement properly when compared to delivery of usual care home visitation services. This belief was rooted in the perception that SafeCare necessitated unyielding adherence to the RSI protocol. That said, the majority of frontline workers in both T1 and T2 described moving from an initial impression of SafeCare as being extremely rigid to the understanding that some degree of flexibility was integral to implementing the model and was, in fact, expected.

Yet, a level of ambiguity persisted through T1 and T2, contributing to anxiety among home visitors and coaches about faithfulness to the RSI model. At the end of T1, several home visitors expressed confusion over whether they could incorporate material from other curricula into SafeCare. In T2, one home visitor stated: “I’m doing [SafeCare], but I don’t feel confident that I’m doing it right.” Home visitors explained that they wanted to follow SafeCare “to a ‘T’” but were simply unsure whether they were doing so. For instance, participants in one lively focus group discussion debated whether they should read the module scripts to clients verbatim, with one home visitor disclosing, “I thought we could summarize. I don’t read directly,” while a second asserted the opposite. Home visitors also suggested that their coaches were not uniform in what they considered to be “fidelity to the model,” and were also not clear about what parts of the curriculum they could adapt or change. In many cases, the solution of the home visitors was to rely on their own interpretations of the curriculum when implementing SafeCare. One seasoned home visitor explained, “I know that I just kind of wing it sometimes. I’m [still] doing the work. I know what I want to see in the client and what the goal is.”

Coaches were also uncertain at times about whether home visitors were implementing SafeCare with fidelity, worrying that monthly fidelity checks might superficially or not fully capture adherence to the model. Coaches in T2 agreed, with one disclosing, “[The home visitors] might be hitting all the points on the [coaching] checklist that we have, but in reality, they’re not meeting fidelity for the session.” Home visitors and coaches commonly reported difficulties measuring and assessing fidelity in the context of their everyday interactions with clients and relied on their own understandings of “doing the work” to meet the ultimate goals of the intervention. However, the act of balancing requirements and measures of both the standard RSI and the research study with their own understandings of fidelity remained a subject of some concern throughout T1 and T2.

Negotiations Within the Inner Context

The outer-context factors described above affected the inner-context experiences of home visitors and supervisors who had to translate and negotiate contract and fidelity requirements while making on-the-ground decisions in order to deliver SafeCare. We describe these complex negotiations below.

Competing Demands and Crises

While home visitors and supervisors enthusiastically opined that SafeCare was a helpful, worthwhile program, the question of how to handle client “crises” (e.g., rent or bills due immediately, lack of housing, and food insecurity) that prevented either implementation or completion of SafeCare, was an ongoing concern in both T1 and T2. According to home visitors, these crises compromised confidence that they were maintaining fidelity to SafeCare (and complying with contracts) when they diverted time away from the modules to attend to what clients were experiencing as more pressing issues. Describing clients in crisis, one home visitor observed, “They’re not really focused or ready, so that kind of puts a strain on you being able to [do SafeCare]. And even if you wanted to help with the crisis, you can’t because of the contract.” Home visitors were concerned that they were not meeting their professional responsibilities as implementers of SafeCare when such situations arose. Yet, they commonly suggested that they were motivated by an equal if not greater responsibility to help clients in crisis address their individual or family needs. While home visitors generally appreciated having a predetermined method for conducting a home visit, they also described ambivalence about adhering to SafeCare in times of crisis. One home visitor commented, “In a way it’s like we wash our hands. We just go back to our SafeCare parenting [module], which is good, but at the same time it’s like there’s no room to just support parents with crisis.” A second home visitor clarified,

Instead of just trying to take care of the problem, there’s a certain sense that I’m obligated to do this [SafeCare]…. I’ve set the curriculum aside a number of times. Sometimes you just have to set the curriculum aside, and then I feel like I’m not meeting some deadline of making progress [toward SafeCare goals]. To some extent it’s personal but every so often we get a chart that we have to update and it brought my attention to how much time is lapsed and how little progress I can say I made.

For these home visitors, the competing demands of SafeCare implementation and individual client needs were experienced as a conflict, both in professional responsibilities, as exemplified in the aforementioned chart illustrating scant client progress, and in personal feelings of competence.

In these situations, home visitors sought to reconcile their own sense of what it meant to truly help clients with their professional obligations as SafeCare practitioners. Their conflicting concerns on this matter commonly remained unresolved, yet some home visitors felt strongly about asserting their ethics of care over perceived contract obligations. For example, one home visitor insisted, “I will never put a family at risk. If they need to talk, I’ll do it. I’ll have to do what I have to do. And then I’ll just work it out later with my supervisor. That’s how I see it.” In sum, ideas and beliefs about contract and fidelity requirements challenged the values of service that were commonly espoused by both home visitors and their CBO supervisors.

Structure versus Creativity

As noted elsewhere, fidelity to the SafeCare model was a topic of much discussion in T1 and T2. Fidelity to this model was not only a work obligation that required reconciling outer-context demands of the intervention developers with inner-context realities, it also represented an issue of personal negotiation for workers that centered on their identities as professionals within the inner context. Home visitors, coaches, and supervisors conveyed a great deal of satisfaction with the model’s structured nature, stating that it gave them a sense of making a difference in clients’ lives. A home visitor in T1 explained, “It is more structure and sometimes most of the clients need more structure. [Parents] have to get it [a module], otherwise, you won’t get to the other module. So you make sure that at least they make a little change. That’s a good thing for you as a provider.”

The structure of SafeCare enhanced feelings of competence and security, especially for those with prior home visitation experience based on other curricula that reportedly fostered reliance on individual discretion. In T2, one home visitor explained: “[With SafeCare], I feel more confident because I have it [the curriculum] there to back me up. All the information is right there on the page. Whenever they [clients] have questions, SafeCare provides me more concrete answers than the other curriculum does.” For home visitors, the structure of SafeCare reportedly enabled them to provide clear, evidence-supported answers to client questions, thus increasing their own sense of expertise as professionals and reducing the possibility of wayward acts of discretion undermining implementation. On the other hand, some workers in both T1 and T2 missed the chance to “think on their feet” that the less-structured services they delivered in the past had required. One individual stated, “Maybe it made me less creative, because I have something solid to go on, [which is] good and bad for me, but I kind of like doing on-the-spot thinking, so like previous experience and all that stuff I bring it in with the previous curriculum, but with this one [SafeCare] I don’t have to. It’s kind of all just laid out for me.” This worker’s ambivalence was reiterated by others who expressed the desire for more “space and responsibility” to improvise, bring in other experiences, and engage in creative problem-solving.

Relationships with Clients

A final area in which home visitors negotiated their own personal and professional values within the inner context of the RSI was in their relationships with clients. As the majority of home visitors, coaches, and supervisors pointed out, the implementation of SafeCare occurred in a sensitive environment in which families were at-risk for, or had documented instances of, child neglect. In this environment, parents might be angry, embarrassed, or resistant to instruction, emotions that strained home visitors’ personal relationships with them. Many home visitors reported discomfort with the need to assert authority over what they perceived as personal and private matters, such as clients’ homes and hygiene practices. This need, they suggested, was reinforced by the RSI’s seemingly top-down approach to imparting parenting knowledge and skills. For example, the SafeCare curriculum covers how to properly read a thermometer and determine how to make health care decisions based, in part, on the reading. Although teaching this content is integral to SafeCare, home visitors suggested that some clients view such material as paternalistic and demeaning, requiring home visitors to broach the topic as tactfully as possible. One individual explained: “You have to be delicate and mindful in the manner in which you present something so elementary without hurting the feelings of the parent that’s receiving this.”

Home visitors tried addressing this awkwardness by alternately identifying with, and distancing themselves from, the RSI model. Several home visitors described situations in which they attempted to diffuse parental concerns of being judged or patronized by explaining that they themselves had learned a lot from SafeCare. One home visitor commented on the Parent-Child Interaction module, “I really like it. I mean I use it myself you know, so it’s really good. So I just let [the parents] know that I use it and it works.” Others made sure that parents knew that they used the SafeCare health manual in their own homes. Several reported using reframing strategies to avoid presenting modules and tasks in a condescending way. One home visitor explained,

I just tell parents that, even if they already know it, it’s a good reinforcer for the agencies and child welfare too because sometimes they don’t see the things we see… I’m like: ‘Look, I know you know this, but you need something behind you that you can point to and say you did’… So I’ll do it anyway so that they’ll have that paper trail behind them.

In this way, home visitors sought to lessen the stigma possibly experienced by parents who were being taught basic skills by explaining that it was just “part of the curriculum” they were required to implement. At the same time, home visitors believed they were able to maintain good personal relationships with clients and avoid being perceived as judgmental or condescending.

Discussion Applications to Practice

This research illustrates how outer- and inner-context factors place complex and sometimes competing demands on frontline service workers and their supervisors in the course of fulfilling their everyday tasks and responsibilities. Our understanding of the use of discretion in coping with competing demands during RSI provision is limited by our focus on frontline workers and supervisors in a single system. Its generalizability is thus constrained. Yet, this study is unique in examining these perspectives of workers and associated demands in depth, and the nuanced negotiations made to accommodate them. It is also distinctive in examining these dynamics over two time periods, T1 and T2, during which a RSI was implemented system-wide. By tracking implementation over time, this study illustrates how the implementation demands on frontline workers shifted. For example, as might be expected, while home visitors and supervisors were extremely concerned by contract changes at the beginning of T1, supervisors reported that things had “mellowed out” during T2. Similarly, apprehensions about accountability within practice circumstances influenced by overlapping levels of higher authority were reported often at T1, but had largely subsided by T2. Finally, anxieties about fidelity among home visitors and supervisors decreased from T1 to T2. This is consistent with a system, organization, and practice culture change that is likely to occur when a new innovation becomes instantiated and sustained (Aarons, Horowitz, Dlugosz, & Ehrhart, 2012; Aarons et al., 2011).

However, this study also highlights several areas in which frontline workers and their supervisors continued to negotiate inner- and outer-context factors well into the implementation of SafeCare. For example, two years after the change in billing requirements for “face-to-face” services occurred at the beginning of T2, home visitors and supervisors were still experiencing conflicts between “quantity” and “quality” of service. Likewise, home visitors continued juggling the exigencies of client crises with the structured practice of Safecare throughout T1 and T2, while also attempting to reconcile RSI implementation with their expressed desire for creativity and sense of appropriate interpersonal relationships with parents. Characteristics of this process were the worries about whether the home visitors delivered the RSI with adequate fidelity, which, while sounding like a concrete, measurable property, can often be difficult to identify and measure in practice (Aarons, Green, Palinkas, Self-Brown, Whitaker, Luktzker… Chaffin, 2012, Schoenwald et al. 2011). In many cases, home visitors found themselves having to balance contractual demands, time pressures, and the individual needs of clients with some uncertainty about the precise meaning of what fidelity entailed. Nonetheless, home visitors praised the specific structure of the RSI and conscientiously tried to resolve this tension as they worked to meet the multifaceted needs of clients.

Other studies have also documented similar demands because placed on frontline workers and supervisors because of shifts in outer-context conditions (Ellis 2011, Noble and Irwin 2009, Willging et al. 2014). However, this study suggests that the need to exercise discretion to negotiate inner- and outer-context factors is not simply a characteristic of RSI implementation processes, but rather an ongoing and dynamic aspect of service delivery deserving of greater appreciation and study. Even after systemic changes are routinized, frontline service workers continue to engage in a process of strategic decision-making in their everyday implementation of the RSI. Similarly, this study indicates that worker tensions to stay true to fidelity within complicated service environments may represent a ubiquitous aspect of service delivery.

Home visitors and their supervisors involved in SafeCare represent the “situated agents” described in scholarship on frontline service work (Durose 2009, P. 36). As such, home visitors and supervisors operated in a nexus in which they worked to reconcile the outer context of interagency politics, contract requirements, and RSI fidelity, with the inner contexts of client crises and demands, along with their own senses of professionalism, ethics of care, and interpersonal relationships with families. As situated agents, these individuals made context-specific judgments about how to best comply with formal requirements in order to bring their day-to-day tasks more in line with their values and conceptions of valuable and meaningful work (Prior and Barnes 2011). They simultaneously strove to maintain good relationships with their clients and colleagues to diffuse tension and to preserve their own feelings of competence, professionalism, and the “proper way to live and to bear responsibility to others” (Brodwin 2012 p. 19). While the street-level bureaucrat “makes” policy through the exercise of discretion (Lipsky 1980), these workers were themselves translating policy into everyday practice and ethics of care. For example, supervisors attempted to buffer home visitors from outer-context influences, such as contract changes, as home visitors crafted ways to accommodate client needs, protect client feelings, and reduce client stigma during the implementation of a structured intervention. In this way, we can expand on Lipsky’s “street-level bureaucracy” to theorize both supervisors and home visitors as not only street-level bureaucrats, but also as street-level mediators and street-level ethicists, whose everyday acts of discretion may likely represent an ongoing aspect of implementation and sustainment of RSIs within human service systems.

The efforts of home visitors and supervisors to reconcile perceived discrepancies in their values with SafeCare implementation can be seen as exercises of accommodation, assimilation, and coping strategies, which Aarons and Palinkas (2007) identify as crucial to job stability and the successful sustainment of RSIs. According to Aarons and Palinkas (2007), service providers must not only adapt an RSI to a local context, they must also adapt to new or changed personal and interpersonal relationships. Clearly, providers of the RSI SafeCare County are attempting to do just this. However, these process of negotiation and accommodation may constitute a burden to frontline workers and their supervisors. For example, home visitors in this study described decisions to work longer hours to prepare for client visits rather than sacrifice time spent with clients. Similarly, supervisors bore the brunt of sorting through the “blasts of e-mails,”, which explained new and conflicting regulations and requirements related to basic administrative issues and the larger research study, to shield their staff from unnecessary confusions. In these examples, we contend that frontline workers and their supervisors were engaging in the unremunerated labors of “committed work” (Hopper 2006 p. 221). Committed work may be the byproduct of RSI implementation in specific contexts; here it was possibly integral to both implementation and sustainment of SafeCare, an intervention that was highly valued by home visitors and supervisors alike. Organizational responses to the reimbursement mechanism introduced in T2 to better compensate CBOs in SafeCare delivery—rather than the RSI itself—may have fostered this reliance on committed work. Heightened organizational expectations and administrative pressures for face-to-face time to increase billable hours likely led to the extra efforts on the part of home visitors to attend to various aspects of RSI delivery that did not involve in-person contact with families, and to also tackle client needs perceived as outside the scope of the RSI. Such committed work requires continued attention, both by researchers and by contract and policy creators and upper-level CBO administrators.

This analysis did not draw from interview or focus group data collected from either public officials or CBO management involved in SafeCare implementation. Such data are discussed at length in separate publications (Aarons et., 2014: Willging, green, Gunderson, Chaffin & Aarons, 2015) and do not center on issues of discretion at the frontlines. Future analyses will involve a systematic comparison of inner-and outer-contextual factors as defined and experienced by differently positioned SafeCare stakeholders (e.g., frontline workers, supervisors, coaches, public officials, and CBO managers). In addition, we did not sample service recipients for this study. Further research must focus on their particular experiences with the RSI in relation to their encounters with frontline workers. This qualitative study would also be enhanced by the use of methods such as ethnography to allow for participant observation of frontline workers and supervisors to corroborate the interview and focus group data.

Despite the study limitations, this study produced understandings of frontline service delivery that have a number of implications for those involved in social work practice and related helping professions at both the frontline and system levels. Notably, the majority of study participants were, in fact, trained in social work. RSIs are considered to be part of ethical social worker practice, aiding in structuring decision making based on available evidence, such as client preferences, available resources, and clinical expertise (Thyer, 2015). However, home visitors in the trenches may benefit from periodic reminders from their supervisors and others involved in implementation about the inherent flexibility that is part of RSIs such as SafeCare (Cohen et al., 2008; Glasgow, Klesges, Dzewaltowski, Bull & Estabroosk, 2004; Rogers, 1995). Dispelling persistent ambiguities about the degree of flexibility allowable in a RSI within the bounds of fidelity may help resolve tensions in the workday experiences of home visitors. Fidelity also needs to e defined and operationalized as a real and meaningful measure of the quality of services that clients are receiving, rather than simply a “rubber stamp for business as usual” or a way to question or police the professional competence of providers (Gambrill, 2006, P. 354). Academic partners have ongoing roles to play in opening channels of communication and providing additional training to quell concerns about the perceived inadequacies of fidelity checks related to SafeCare and other RSIs. Communication fostered through both development and use of reliable feedback mechanisms that engage frontline workers in disucssions with academic partners about fidelity, and in other monitoring and evaluation activities, can be useful for identifying further training needs.

In addition to the academic partners, organizations- and system-level stakeholders also have pivotal roles to play in opening lines of communication. Supportive coaches, for example, can proactively mediate between the experiences of frontline workers and supervisors and the fidelity requirements of SafeCare and other RSIs. Both CBO administrators and government funders who formulate and sign off on contracts may need to become more hands-on in consulting with home visitors about the administrative requirements that shape SafeCare provision. Our findings identify sufficient funding and time for administrative work as key factors to consider in future contracts to facilitate RSI implementation and sustainment. Administrators can also ensure that home visitors can avail themselves of continued opportunities for professional development. Such measures may go a long way in alleviating concerns about fidelity and increasing or maintaining buy-in for RSIs (Aarons, et al., 2011).

The significance of provider values in social work practice cannot be overstated. Values are not merely manifestations of “intrusion or noise” to purge from RSI delivery but are ever-present phenomena requiring greater study and appreciation (Longhofer & Floersch, 2013. P. 527). A focus on the values or ethics of care that social workers enact through their work can be used to formulate research questions pertinent to RSI delivery and its overall enhancement. For example, social justice is a value often espoused by social workers. Therefore, research within the context of RSI implementation can explore how social workers enact a social justice orientation in their work, given the perceived constraints that a service delivery model may place on their practice (Longhofer & Floersch, 2013). A qualitative approach can identify areas where frontline workers experience misalignments between how they understand administrative requirements built into contracts and therefore CBO workplace operations, and how they implement RSIs in practice while maintaining their professional identities and ethics of care. A qualitative approach also offers insight into the kinds of negotiations, accommodations, and coping strategies that they may employ to resolve these misalignments. Such an approach promotes broader awareness of the committed work that is performed to translate outer context demands into everyday practice, and unintended consequences of contracts and policies that shape the environments in which frontline workers exercise discretion.

Finally, although RSIs by their nature tend to be implemented in an authoritative, top-down approach, we argue that it still remains incredibly important to tap into the potential for bottom-up change based on social worker perspectives (Gambrill, 2006). Although RSIs are intended to respond to the uncertain nature of real-life decision-making situations (Gibbs & Gambrill, 2002), the specific knowledge of and experiences of local contextual factors (both inner and outer) of social workers at the frontlines can lead to important adjustments to RSI implementation strategies and environments that may reduce the everyday tensions and conflicts described in this study, thereby better aligning values of practice with the provision of quality services (Gambrill, 2006). We contend that it is vital to honor the perspectives of social workers at the frontlines, drawing from their situated vantage points to inform the content and conduct of RSIs and other interventions.