Evidence-based practice adaptation during large-scale implementation: a taxonomy of process and content adaptations
| Published date | 06 June 2019 |
| Date | 06 June 2019 |
| DOI | https://doi.org/10.1108/JCS-02-2018-0003 |
| Pages | 61-77 |
| Author | Gregory A. Aarons,Rachel A. Askew,Amy E. Green,Alexis J. Yalon,Kendal Reeder,Lawrence A. Palinkas |
| Subject Matter | Health & social care |
Evidence-based practice adaptation during
large-scale implementation: a taxonomy of
process and content adaptations
Gregory A. Aarons, Rachel A. Askew, Amy E. Green, Alexis J. Yalon, Kendal Reeder and
Lawrence A. Palinkas
Abstract
Purpose –The purpose of this paper is twofold: first, to identify the types of adaptations made by service
providers(i.e. practitioners) duringa large-scaleUS statewide implementation of SafeCare®, an evidence-based
intervention to reducechild neglect;and second, to placeadaptationswithin a taxonomyof types of adaptations.
Design/methodology/approach –Semi-structured interviews and focus groups were conducted with 138
SafeCare providers and supervisors. Grounded theory methods were used to identify themes, specific types
of adaptations and factors associated with adaptation.
Findings –Adaptations were made to both peripheral and core elements of the evidence-based practice
(EBP). The taxonomy of adaptations included two broad categories of process and content. Process
adaptations included presentation of materials, dosage/intensity of sessions, order of presentation,
addressing urgent concerns before focusing on the EBP and supplementing information to model materials.
Content adaptations included excluding parts of the EBP and overemphasizing certain aspects of the EBP.
Adaptations were motivated by client factors such as the age of the target child, provider factors such as a
providers’level of self-efficacy with the EBP and concerns over client/provider rapport. Client factors were
paramount in motivating adaptations of all kinds.
Research limitations/implications –The present findings highlight the need to examine ways in which
adaptationsaffect EBP implementation and sustainment, clientengagement in treatment, and clientoutcomes.
Practical implications –Implementers and EBP developers and trainers should build flexibility into their
models while safeguarding core intervention elements that drive positive client outcomes.
Originality/value –This study is unique in examining and enumerating both process and content types of
adaptations in a large-scale child neglect implementation study. In addition, such adaptations may be
generalizable to other types of EBPs.
Keywords Adaptation, Evidence-based practice, Implementation, Child maltreatment, Child neglect,
Empirically supported treatment, Child welfare
Paper type Research paper
As evidence-based practices (EBPs) move out of controlled settings, such as efficacy studies,
and begin to be implemented in real-world community settings, a tension often arises between
strict adherence to the original model and adaptations based on local circumstances and needs
(Backer, 2001; Bauman et al., 1991). Research on implementing EBPs in community settings
suggests that some local adaptation may be necessary in order to ensure that the intervention
fits the needs of subsets of service users (Lau, 2006). That is, having some flexibility built into an
EBP may be important to client engagement and retention in services and successful client
outcomes (Barrera and Castro, 2006; Lau, 2006; McHugh et al., 2009).
Recent literature has made the case for and described efforts to address planned adaptations as
a part of the implementation process (Aarons, Green, Palinkas, Self-Brown, Whitaker, Lutzker
and Chaffin, 2012; Aarons, Miller, Green, Perrott and Bradway, 2012; Barrera and Castro, 2006;
Cabassa and Baumann, 2013; Lau, 2006; Lee et al., 2008; Solomon et al., 2006; Wandersman,
2003). Adaptations may be seen as necessary to address the unique needs of clients or patients
Received 23 February 2018
Revised 5 July 2018
2 February 2019
Accepted 14 March 2019
(Information about the authors
can be found at the end of
this article.)
DOI 10.1108/JCS-02-2018-0003 VOL. 14 NO. 2 2019, pp. 61-77, © Emerald Publishing Limited, ISSN 1746-6660
j
JOURNAL OF CHILDREN'S SERVICES
j
PAG E 61
(Kakeeto et al., 2017); in one study, culturally adapted child and youth preventative programs
were found to be even more effective than programs that were not adapted (Hasson et al., 2014).
However, the majority of adaptations made to EBPs are likely unplanned and less deliberated and
structured than those derived from “planned adaptations”models discussed in the literature
(Aarons, Green, Palinkas, Self-Brown, Whitaker, Lutzker and Chaffin, 2012; Elliott and Mihalic,
2004; Southam-Gerow, 2004). While the literature suggests that adaptations to EBPs are
commonplace in real-world implementations (Luongo, 2007; Ringwalt et al., 2004; Rogers et al.,
2005), empirical studies documenting the extent, types and reasons for naturally occurring
adaptations to EBPs are scarce (but see Aarons, Miller, Green, Perrott and Bradway, 2012;
Harshbarger et al., 2006; Hill et al., 2007). Furthermore, while there is widespread agreement that
adaptations to EBPs should not affect the core components or elements of an intervention
(i.e. those required elements that fundamentally define its nature and likely produce the
intervention’s effects) (Cardona et al., 2009; Harshbarger et al., 2006; McKleroy et al., 2006;
Veniegas et al., 2009), literature is lacking about the types and proportion of adaptations made to
existing EBPs that preserve or threaten model fidelity.
The present study addresses these gaps by examining EBP adaptations in the context of a
large-scale implementation of an EBP targeting child neglect, SafeCare®. The SafeCare model,
originally known as Project 12-Ways, is a manualized, structured behavioral-skills training model
delivered to families at risk for child maltreatment. SafeCare is a home-based parenting program
that focuses on helping parents develop problem solving and communication skills and includes
three substantive core modules (Edwards and Lutzker, 2008; Gershater-Molko et al., 2002,
2003): Parent–Child/Infant Interactions, Home Safety and Health. A key part of the SafeCare
model is the use of frequent metrics for documenting parents’starting points and progress over
time in each of the three modules. SafeCare also allows for some flexibility. For example, order of
delivery of the modules may be decided by the provider (i.e. individual practitioner or clinician)
and/or supervisor. Thus, adaptation must be considered in light of EBP requirements. In this
study, we address the following questions:
RQ1. What are the broad categories of adaptations made by providers?
RQ2. What specific types of adaptations do providers make to social service EBPs?
RQ3. What are providers’motivations for making adaptations?
Methods
Study context
Thestudytookplaceacrossanentirechildwelfare system in one large state within the USA. In
collaboration with academic researchers, the state child welfare system selected SafeCare as
an EBP to implement th rough the state chi ld welfare family pr eservation/fami ly reunification
service system. Community-based organizations were already contracted with the child welfare
system to provide home-based services to families identified and/or at risk for child abuse or
neglect. Thus, Saf eCare was adopted and used as the prima ry service model for home-based
services in this children’s system and with fami lies across a range of ri sk levels. Most familie s
receiving SafeCare were referred either for preventative services or in conjunction with a
court-mandated treatment plan. The effectiveness of SafeCare and the implementation
process was studied via an effectiveness trial focused on client outcomes (Chaffin et al., 2012)
and a mixed methods implementation study focused on factors that facilitate or impede EBP
implementation in a real-world child ren’s services setting (Aarons and Sommerfeld, 2012;
Aarons, Sommerfe ld, Hecht, Silovsky and Chaffin, 2009). The effectiveness trial used a 2 (EBP
vs services as usual) ×2 (fidelity mo nitoring vs no fidelity monito ring) study design. All provide rs
received training in the same EBP. Due to the number of providers and the experimental design
of the effectivene ss trial, trainings were do ne based on region in light of pr actical concerns such
as scheduling and lo cation. Supervis ors were not required to be trained as their role was
primarily in overseeing scheduling and management of their teams. Coaching in the model was
provided by university-based trainers and coaches. Coaching and feedback was provided
monthly through in v ivo observation and feedback. The present study utilized qualitative data
PAG E 62
j
JOURNAL OF CHILDREN'S SERVICES
j
VOL. 14 NO. 2 2019
To continue reading
Request your trial