Souraya Sidani - Nursing and Health Interventions

Recently, concerns have been raised about the applicability of evidence, derived from primary and meta‐analytic studies using the RCT design, in informing practice (Ioannidis, 2016). Overall, the concerns stem from limitations of the RCT design in generating evidence that is relevant to the practice context (Braithwaite et al., 2018; Reeve et al., 2016). The limitations are related to the features of the RCT (i.e. careful selection of participants, random assignment, standardized delivery of treatment) that enable the focus on the direct causal effects of an intervention on outcomes and the control of potential sources of bias. As such, the RCT features ignore the complexity of the real world, the individuality of clients' experiences of the health problem and life circumstances as well as responses to treatment, and clients' participation in treatment decisions.

Careful selection of participants confines the RCT sample to a select subgroup of the target client population (e.g. clients with no comorbid conditions), which limits the applicability of the findings to other subgroups of clients seen in practice (Greenhalgh et al., 2014). Random assignment of participants to treatment groups does not reflect the treatment decision‐making process followed in practice. Therefore, random assignment is not well received by clients participating in the RCT (thereafter referred to as participants) and has been found to affect enrollment in the trial, attrition and nonadherence to treatment, which weaken the validity of inferences regarding the effectiveness of an intervention (see Chapter 14). Standardized delivery of interventions is not responsive to clients' individual experiences, life circumstances, and preferences. Standardization also is difficult to transport into practice due to the complex and inter‐related influence of factors pertaining to clients, health professionals, and context (Chu & Leino, 2017; Leask et al., 2019). The focus on the average direct causal effects of the intervention ignores individual variability in clients' responses to treatment (i.e. level of improvement in outcomes observed following treatment completion) and the mechanism through which the treatment produces its benefits; yet, health professionals need to understand what client subgroups respond favorably to the intervention and how the intervention produces its benefits for making appropriate treatment decisions (Horwitz et al., 2017; Lipsitz & Markowitz, 2013; Van Belle et al., 2016).

The limitations extend to meta‐analytic studies or systematic reviews of RCT findings, which form the basis for recommendations stated in guidelines. Attempts at synthesizing RCT‐derived evidence face challenges associated with limited replication (e.g. Pereira & Ioannidis, 2011). Limited replication is manifested in conflicting and, therefore, inconclusive evidence of the intervention's effectiveness (Hesselink et al., 2014). Accordingly, the guidelines' recommendations are usually stated in general terms that simply identify the interventions that can be used in addressing a health problem (Edwards et al., 2007). In addition, reports of primary and meta‐analytic studies as well as guidelines provide a brief description of the interventions. Insufficient description of the interventions constrains their replication and proper implementation in research and practice (Bach‐Mortensen et al., 2018; Levinton, 2017). For instance, Glasziou et al. (2010) found that health professionals were able to replicate the interventions evaluated in half of 80 studies published in the journal of Evidence‐Based Medicine. Furthermore, the guidelines do not offer instructions on how to adapt the design and delivery of interventions in a way that preserves their active ingredients yet is responsive to the characteristics, preferences, and life circumstances of clients and to the resources available in local practice contexts (Bach‐Mortensen et al., 2018; Westfall et al., 2009).

Accordingly, the evidence generated in primary and meta‐analytic studies using the RCT design is of limited utility in informing practice. It does not address the questions that health professionals ask when making treatment decisions (Bonell et al., 2018; Levinton, 2017). The questions include:

Who (i.e. clients with what sociodemographic and health or clinical profiles) most benefit (i.e. demonstrate improvement in outcomes) from an intervention, delivered in what mode and at what dose?

What are the intervention's active ingredients (operationalized in what specific components) responsible for its benefits?

What risks or discomforts are associated with the intervention?

How and why does the intervention work to produce its benefits? Or, what is the mechanism of action responsible for the intervention's effectiveness in addressing the health problem?

What resources are needed to deliver the intervention?

What contextual factors influence the delivery of the intervention by health professionals, its uptake and enactment by clients, and its effectiveness?

To what extent and how can the intervention be tailored to the individual clients' characteristics or preferences, and/or adapted to the local practice context?

What alternative interventions are available to address the health problem, and what are their relative benefits (effectiveness) and risks (safety)?

Intervention research needs to be reoriented toward developing well‐conceptualized yet practice‐relevant interventions, and generating the evidence that addresses these questions. The goal is to consolidate the theoretical and empirical knowledge that informs practice, and ultimately improves the quality of healthcare and the health of clients. To be useful in informing practice, intervention research should embrace a realist, pragmatic perspective in reflecting the characteristics of practice: client‐centeredness and complexity. This can be achieved through client engagement and use of a range of relevant research designs and methods.

The less‐than‐optimal experiences with evidence‐based practice, the limited applicability of RCT‐derived evidence to practice, in combination with clients' demand for an approach to healthcare that reflects their individuality, values, and preference, have led to the resurgence of client‐centered care as the “core” of high‐quality healthcare (Beck et al., 2010; de Boer et al., 2013; Sidani & Fox, 2014; Van Belle et al., 2019; Vijn et al., 2018).

Client‐centeredness is an approach to healthcare familiar to health professionals. Professionals are instructed, socialized, and expected to deliver client‐centered care. Client‐centered care is applied at different levels. At the individual level, it involves the application of tailored and adaptive interventions addressing the presenting health problem or aiming to change health behaviors (Hekler et al., 2018) and personalized or precision medicine (Bothwell et al., 2016). At the group level, client‐centeredness is illustrated by family‐centered care or the provision of health interventions that are adapted to the demands and preferences of particular communities such as ethno‐cultural communities (Barrera et al., 2013; Netto et al., 2010). At the healthcare organization level, client‐centeredness involves the adaptation of evidence‐based interventions and practice guidelines to the local context (Harrison et al., 2010; Powell et al., 2017) and at the system level, it is reflected in patient engagement (McNeil et al., 2016).

In general, the application of client‐centered care involves: (1) a comprehensive and thorough assessment of the clients' condition to identify their health problems, beliefs, values and preferences; (2) collaboration and active participation of clients in prioritizing their problems, designing new or selecting available, evidence‐based interventions, and implementing the selected interventions (as is done in shared decision‐making); and (3) adaptation or tailoring of the intervention for consistency with clients' problems, beliefs, values, and preferences, as well as with their changing experiences of the health problem, and life circumstances, over time.