Reflection: A Middle-Out Approach to Implementing eHealth Innovation

Upon reflecting of the different innovation processes, I’ve asked myself whether innovation should be catalysed and sustain by government actors or agencies or by healthcare professionals and organisations at the grassroots level. Literature examining national information systems from the United Kingdom, the United States, Australia and New Zealand point us towards a middle-out approach to implementing eHealth innovations.

The United Kingdom’s National Programme for Information Technology’s (NPfIT) deployment of a general EHR system has uncovered distinct shortfalls of nationally mandated EHRs (Coiera, 2009). While the NPfIT has demonstrated the importance of national-scale information systems (NHISs) being built on a pervasive information structure, the multiple “setbacks, misgivings, clinical unrest, delays, cost overruns, and paring back of promised functionality” resulting in calls for the program to be shut down highlight the complexities of governments taking a leading role in the implementation of Health Information technology (HIT) (Coiera, 2009). The standard nature of these EHRs are not flexible enough to be tailored for their specific healthcare uses. The highly structured and content constrained nature of national EHR system makes it difficult for healthcare providers in a pathway or providers across different organisational boundaries to use the same system to share patient information (Eason et al., 2012). These national EHR systems also fail to accommodate the progressive development of local innovations, they are characterised by a static design rather than continuous evolution. Additionally, national systems are often implemented with national procedure for data governance that may not align with realities in local context (e.g. role-based access rules) (Eason et al., 2012).

On the other hand, the United States’ highly fragmented health system has attempted to apply a bottom-up approach to developing their health information infrastructure. This has culminated in local health providers forming regional coalition to “interconnect their existing systems” into amalgamated health information exchanges (HIE). The hope is that regional HIEs will be able to eventually aggregate into a nation-wide health information system. Both the United States’ and UK’s national health systems hope to achieve the same goal: EHRs that are accessible and shared by different healthcare providers that can add or read information from others (Coiera, 2009). Bottom-out approaches offer pragmatism and cost efficiency. Preserving existing local EHR systems avoids the costs of replacing them with new systems and retraining staff. The risk however is that the extent of interoperability of these systems is arbitrary, resulting in a weaker national health information system. Additionally, compliance with standards, governance of patient data and alignment with national goals can be a voluntary activity by local health providers (Coiera, 2009).

In the face of “one size does not fit all” in regard to national EHR system and the impracticality of a plethora of locally designed eHealth innovations that would be inoperable between different healthcare providers, there is an opportunity for mid-level or middle-out solutions that find the functional values of local healthcare providers and the governance values of national actors (Eason et al., 2012).

Several commentators have suggested the need for systems development have proximity to where clinical services are delivered (Eason et al., 2012). Likewise, Sheikh et al., (2021) argue that system-level changes need to be responsive to the needs and perspectives of multiple stakeholders, as well as the “political, regulatory and sociocultural context of health systems”. This is also reflective of Health Information Technology's (HIT) requirement for mutual adoption at both the facility level and the government level so that technology and work processes can be aligned. HIT implementation is a complex activity that needs to be managed adaptively to accommodate local contexts (Sheikh et al., 2021). Thus, these considerations when applying any innovations in a health system, particularly technological innovations at the meso-level of system design, brings us to the utility of the middle-out approach (Eason et al., 2012).

The middle-out approach is a combination of bottom-up innovation that addresses challenges faced by patients and frontline healthcare professionals/providers, and top-down strategies from the government that implement standard procedures across the entire healthcare system. We can begin to imagine how this could solve the NPfIT's failure to adequately engage the core needs of local healthcare providers that resulted in unattainable expectations in a short turnover and the United States fragmented and inoperable local information health systems.

My research has lead me to two different interpretations of a middle-out approach, but both are complementary. Kranzler, Parag, & Davidovitch's (2019) view of the middle-out approach is as a socio-technical analytical and how-centric perspective that aims to clarify and emphasise “the role of middle actors - stakeholders who are positioned between policymakers and grassroots – as agents of change”. Middle-out actors steer long-term, multi-sectoral processes such as policy design, funding and enabling programmatic continuity.

Top-Down action reflects the capacity to sign legislation, allocate macro-level resources and utilise “political bully pulpits”. In contrast, Bottom-Up action infers its influence through legitimacy and real-world experiences at the grassroots level which is diffused upwards to actors positioned higher in the organisational hierarchy. Middle actors are defined as networks of organised citizen groups, professional associations, faith-based organizations and other institutionalized agents of change who are located between the Top and Bottom. (Kranzler, Parag, & Davidovitch, 2019) add mid-level public servants (from government agencies or local authorities) who “represent, advise and carry out decisions by higher-level policymakers (Top Actors), while also engaging with, seeking advice from and coordinating between individuals and organizations at the grassroots level (Bottom Actors)”. Middle actors and their contributions are often overlooked because they are nether the health intervention target group nor the “policy signatory”. Middle actors specific and defined institutions and organisational structures; specific membership; “official or unofficial procedures or rules; access to unique material resources such as funding and equipment; established communication channels with their members and other Bottom, Middle and Top Actors; as well as non-material resources such as professional, spiritual or ethical legitimacies”. These characteristics enable middle actors to exercise authority and agency to “connect, represent, negotiate” and take decisions (Kranzler, Parag, & Davidovitch, 2019).

The middle-out approach utilises “in-betweenness as a space” from which middle-out actors integrate their strong familiarity with and accessibility to policies, programmes, procedures and stakeholders that shape actors/entities at the individual/community level. In a public health context, middle-out actors occupy the organisational/institutional dimension of the Social-Ecological Model, which place them in between the interpersonal and community dimension (Kranzler, Parag, & Davidovitch, 2019).

Their placement between decision-makers and local health communities enable middle-out actors to multi-directional influence such as impacting policies, strengthening grassroots initiatives and facilitating inter-agency. Inter-agency is expressed as a Sideways influence in their model and includes all activities that create and sustain cooperation across grassroots actors, decision-makers and departmental, organisational and sectoral boundaries (Kranzler, Parag, & Davidovitch, 2019).

Rather than the roles and responsibilities of actors who occupy the meso-level in a health system, the second interpretation of the middle-out approach recognises that governments and providers have different starting points and resources (degree of preparedness of eHealth innovation) as well as goals. The earlier stages are characterised by governments primarily focus on helping to fund, support and incentivise the adoption of technically and functionally compliant technologies at the local level “rather than setting mandates”.

A middle-out approach would enable actors at the macro-level to co-design standards and governance procedures alongside local actors through the innovation of EHR system. This approach will require intimate partnerships, value and priority setting and “interagency between the two levels (Eason et al., 2012). This could realize a set of shared goals and guiding standards between the two groups. This propels providers to converge to similar standards while avoiding setting unrealistic or harmful deadlines that could impede adoption (Sheikh et al., 2021). Coiera (2009) proposes that government can take a lead role industries that are immature or where there is strong national interest. For example Australia’s government agency , the national E-health transition authority (NEHTA) has invested into developing interoperability standards well before the contemplating any HIT innovations. Coiera (2009) argues that another compelling element of a middle-out approach is that national health systems that are currently pursuing top-down or bottom-up approaches can migrate to this approach at ant time.

---

References

Coiera, E. (2009). Building a National Health IT System from the Middle Out. Journal of the American Medical Informatics Association, 16(3), 271-273. https://doi.org/10.1197/jamia.M3183

Eason, K., Dent, M., Waterson, P., Tutt, D., & Thornett, A. (2012). Bottom-up and middle-out approaches to electronic patient information systems: a focus on healthcare pathways. Informatics in primary care, 20(1).

Kranzler, Y., Parag, Y., & Davidovitch, N. (2019). Public Health from the Middle-Out: A New Analytical Perspective. International Journal of Environmental Research and Public Health, 16(24), 4993. https://www.mdpi.com/1660-4601/16/24/4993

Krause-Jüttler, G., Weitz, J., & Bork, U. (2022). Interdisciplinary Collaborations in Digital Health Research: Mixed Methods Case Study. JMIR Hum Factors, 9(2), e36579. https://doi.org/10.2196/36579 

Lepore, D., Dolui, K., Tomashchuk, O., Shim, H., Puri, C., Li, Y., Chen, N., & Spigarelli, F. (2023). Interdisciplinary research unlocking innovative solutions in healthcare. Technovation, 120, 102511.

Sheikh, A., Anderson, M., Albala, S., Casadei, B., Franklin, B. D., Richards, M., Taylor, D., Tibble, H., & Mossialos, E. (2021). Health information technology and digital innovation for national learning health and care systems. The Lancet Digital Health, 3(6), e383-e396.