The Process We Think We Have vs. The Process We Actually Use

In healthcare, the distance between policy and practice is rarely zero. Standard operating procedures are written, approved, and filed. Yet on the floor, in the pharmacy, and at the bedside, a different set of steps often unfolds. The process that exists on paper and the process that actually keeps patients safe are frequently not the same thing. Understanding this gap is not an indictment of staff; it is an invitation to look more honestly at how care is truly delivered.

Process mapping is one of the most powerful tools available for this kind of honest examination. When it moves beyond simply drawing the official workflow and instead captures what workers actually do, process mapping can reveal hidden complexity, expose informal workarounds, and surface hazards that would otherwise remain invisible. This article explores why the gap between work-as-imagined and work-as-done matters, how it forms, and what healthcare teams can do to bridge it.

Work-as-Imagined vs. Work-as-Done

The distinction between work-as-imagined (WAI) and work-as-done (WAD) is foundational to the field of Safety II, a framework developed by Erik Hollnagel and colleagues that shifts focus from managing failures to understanding how things usually go right (Hollnagel et al., 2015). Work-as-imagined is the version of a task captured in protocols, training materials, and job descriptions. It reflects what designers, managers, and regulators believe happens during care delivery. Work-as-done is what clinicians and staff actually do when faced with the full complexity of a real working environment.

These two versions of reality diverge for many reasons. Protocols are written under ideal conditions and cannot anticipate every local constraint, resource limitation, or patient-specific variable. Staff adapt continuously, often in ways that improve efficiency and safety, but those adaptations are rarely documented (Braithwaite et al., 2016). The result is a persistent and largely invisible gap between the official process and the lived one.

This gap is not a new observation. Research in human factors and patient safety has documented it across a wide range of clinical settings. A study by Tucker and Edmondson (2003) found that frontline nurses routinely worked around system failures, including missing supplies and incomplete information, at an average rate of one workaround per hour during a typical shift. Importantly, most of these adaptations resolved the immediate problem but did not address its underlying cause, creating the conditions for recurring failures. More recently, Debono et al. (2013) conducted a systematic review confirming that workarounds are ubiquitous in healthcare, driven by time pressure, equipment limitations, and gaps in electronic health record design.

Why Workarounds Emerge

It is tempting to view workarounds as the product of individual carelessness or a culture of cutting corners. The evidence does not support this interpretation. Workarounds typically emerge from a rational response to a system that does not function as intended (Halbesleben et al., 2010). When a nurse bypasses a bar-code medication administration alert because the scanner is broken, or when a physician documents a diagnosis code differently to avoid an insurance denial, these are adaptive behaviors in response to systemic friction.

Halbesleben et al. (2010) proposed a framework identifying three primary drivers of workaround behavior in healthcare: technology barriers (e.g., poorly designed electronic health records), patient-specific demands (e.g., a deteriorating patient requiring immediate action that the protocol does not accommodate), and workflow interruptions. Each driver places pressure on the worker to find an alternative path. In many cases, the workaround works. The patient gets the medication. The note gets written. The problem appears resolved. But the underlying system flaw remains untouched, and the same pressure will generate the same workaround again tomorrow, by a different person, possibly with a different outcome.

This is why workarounds are both a symptom of hidden system dysfunction and a source of new risk. When they become normalized, they may displace the safer, intended process entirely. New staff may learn the workaround as if it were the standard. The institutional memory of why the original process existed begins to erode (Ferneley & Sobreperez, 2006).

The Hidden Complexity of Clinical Work

One of the most consistent findings in healthcare human factors research is that clinical work is far more complex than it appears from the outside, or from above (Carayon et al., 2006). A seemingly simple task, such as administering a medication, may involve navigating multiple software systems, resolving conflicting information from different sources, coordinating with colleagues across roles, managing interruptions, and making real-time judgment calls about patient status. Each step carries decision points that are invisible in the formal protocol.

The Systems Engineering Initiative for Patient Safety (SEIPS) model, developed by Carayon and colleagues (2006), frames healthcare work as the product of an interacting system of people, tasks, tools, environment, and organizational conditions. This model has been widely adopted in patient safety research because it captures the reality that no single element of a care process can be understood in isolation. When a pharmacist checks a medication order, they are working within the constraints of their physical environment, the design of the dispensing system, the information available to them, the time pressure they face, and the norms of their team. The formal process map captures none of this context.

Hidden complexity also accumulates through what Reason (1990) described as latent conditions: system-level factors that are not immediately visible but that set the stage for errors. A cluttered medication room, an ambiguous order format, inadequate handoff communication, and understaffing are all latent conditions. They do not cause errors directly, but they make errors more likely when combined with a triggering event. These conditions can persist undetected for years because no single incident makes them visible. Only when the actual workflow is examined closely do they begin to surface.

Why Traditional Process Maps Fall Short

Standard process mapping produces a diagram of how a process is supposed to work. It is a useful starting point, but it has significant limitations when the goal is to understand how hazards arise. The most fundamental limitation is that it represents work-as-imagined rather than work-as-done.

Hollnagel (2012) introduced the concept of the Functional Resonance Analysis Method (FRAM) specifically to address this problem. FRAM is designed to map the variability in how functions are actually performed, including the upstream and downstream dependencies between them. Because it focuses on what people do rather than what they are supposed to do, FRAM often reveals coupling between tasks that would never appear in a standard flowchart. When one function varies unexpectedly, that variability can resonate through the system, amplifying risk at points far removed from the original deviation.

Traditional process maps also tend to represent linear sequences of steps, which misrepresents the reality of clinical work. Research on workflow interruptions in hospital settings has found that nurses are interrupted or distracted on average every three to five minutes during medication administration, a task that the protocol treats as a discrete, uninterrupted sequence (Westbrook et al., 2010). The gap between the flowchart and reality is not incidental. It is structural.

Mapping What Actually Happens: Methods and Principles

Effective process mapping in healthcare requires methods that prioritize observation and worker input over document review alone. Several approaches have demonstrated value.

Direct Observation

Direct observation involves watching workers perform tasks in real time and recording what actually happens. This approach reliably surfaces deviations from protocol, workarounds, and informal communication patterns that no document would reveal. Observational studies of medication administration processes, for example, have repeatedly found error rates far higher than would be predicted from incident report data alone (Bates et al., 1995). What gets seen from the outside is only what gets reported. Direct observation removes that filter.

Cognitive Task Analysis

Cognitive task analysis (CTA) is a family of methods designed to surface the knowledge, decision-making, and judgment involved in complex tasks (Crandall et al., 2006). Unlike traditional task analysis, which describes physical actions, CTA captures the mental work behind those actions: what information workers are attending to, what they are trying to achieve at each step, and how they recognize when something has gone wrong. This is particularly valuable in healthcare, where much of the hazard lies not in what is done but in how a situation is interpreted.

Participatory Mapping

Participatory process mapping engages frontline workers directly in the mapping exercise. Rather than having analysts draw the process from the outside, staff are asked to describe, and often draw, how they actually do their work. This method consistently produces richer and more accurate maps than externally developed ones, and it has the added benefit of building staff engagement in improvement work (Nicolini et al., 2011). Workers who feel that their experience has been heard are more likely to participate in subsequent changes.

Gap Analysis Between WAI and WAD

Once both the official and actual processes have been documented, a structured gap analysis can identify where they diverge and why. This analysis can distinguish between adaptive deviations that enhance safety and those that introduce new risk. Not every deviation from protocol is dangerous; some represent local learning that the organization would benefit from incorporating into its formal processes (Braithwaite et al., 2016).

What Becomes Visible When You Map Reality

The practical value of work-as-done process mapping is that it makes previously invisible hazards visible. Several patterns tend to emerge consistently across healthcare settings.

• Communication failures at handoffs: Official handoff protocols often assume a structured, uninterrupted transfer of information. Observation routinely reveals incomplete handoffs, missing data, and ambiguous verbal communication that the protocol does not account for (Horwitz et al., 2009).
• Unacknowledged task sharing: In practice, tasks assigned to one role are frequently performed by another. When this informal sharing is not recognized, training, accountability, and error response become misaligned with reality.
• Double documentation: When electronic and paper records are both maintained, staff often develop informal rules about which system gets updated when, creating inconsistencies that can lead to clinical error.
• Alert fatigue adaptations: Staff develop personal filtering strategies for managing high volumes of electronic alerts. These strategies are rarely documented and vary between individuals, creating unpredictable patterns of response (van der Sijs et al., 2006).
• Environmental workarounds: Physical layout problems, such as medication storage that requires multiple trips across a unit, generate consistent workarounds that introduce handling errors.

Each of these patterns represents a hazard that would not appear in a standard process review. They become visible only through systematic examination of actual work.

From Mapping to Improvement

Identifying the gap between work-as-imagined and work-as-done is not an endpoint; it is a starting point. Once actual workflows have been mapped and hazards identified, organizations face a choice about how to respond.

One option is to enforce the official process more strictly, treating the gap as a compliance problem. This approach has poor evidence behind it. When workers deviate from protocols because those protocols do not fit their actual conditions, stricter enforcement typically produces resentment, increased workarounds, and underreporting of problems (Braithwaite et al., 2016). It does not address the underlying system design issues that made the deviation necessary.

A more productive approach is to use the mapping findings to redesign processes so that the intended workflow is also the easiest and most natural one to follow. This principle, sometimes described as making the safe path the path of least resistance, has been applied successfully in medication safety, surgical checklists, and handoff redesign (Pronovost et al., 2006). It treats the workaround not as a problem to be eliminated but as a signal about where the system needs to change.

Organizations may also find that some informal adaptations represent genuine improvements that should be formalized. Staff who have developed better ways of performing a task, even if outside the official protocol, are a source of innovation. Participatory mapping creates an opportunity to surface these improvements and incorporate them into updated standards.

Implications for Healthcare Teams

For healthcare workers, the core message of this literature is both challenging and affirming. It is challenging because it requires honesty about the gap between official processes and actual ones, and that honesty can feel threatening in an environment where deviation from protocol is often treated as an individual failure. It is affirming because it locates many of the root causes of error in system design rather than individual behavior, and it respects the adaptive intelligence that frontline workers bring to their work every day.

Several practical principles follow from the evidence.

• Do not assume that the written protocol reflects how work is actually done. Ask workers to describe and demonstrate their actual steps, not the official ones.
• Treat workarounds as information. When staff consistently find ways around a step in the process, the process likely needs to change.
• Create psychological safety for honest mapping. Workers will not describe their actual workflow if they fear that deviations will be used against them.
• Involve frontline staff in redesign. The people who know where the friction is in a process are the best positioned to identify what would make it better.
• Revisit maps regularly. Workflows change over time, especially with new technology or staffing changes, and the gap between WAI and WAD can widen silently.

Conclusion

The process we think we have and the process we actually use are rarely the same. This gap is not evidence of failure; it is evidence of complexity. Healthcare workers adapt constantly to a system that is imperfect, resource-constrained, and often poorly designed for the work it demands. Those adaptations are sometimes brilliant and sometimes hazardous, and they are almost always invisible in official documentation.

Meaningful process mapping requires the courage to look at what is actually happening rather than what should be happening. When organizations make that investment, they consistently find hazards that conventional approaches missed, and they find opportunities for improvement that no policy review would have surfaced. The map that reflects reality is not a more embarrassing document than the idealized one. It is a more useful one.

References

Bates, D. W., Cullen, D. J., Laird, N., Petersen, L. A., Small, S. D., Servi, D., Laffel, G., Sweitzer, B. J., Shea, B. F., Hallisey, R., Vander Vliet, M., Nemeskal, R., & Leape, L. L. (1995). Incidence of adverse drug events and potential adverse drug events: Implications for prevention. JAMA, 274(1), 29-34. https://doi.org/10.1001/jama.1995.03530010043033

Braithwaite, J., Wears, R. L., & Hollnagel, E. (2016). Resilient health care: Reconciling work-as-imagined and work-as-done. In J. Braithwaite, R. L. Wears, & E. Hollnagel (Eds.), Resilient Health Care, Vol. 3: Reconciling Work-As-Imagined and Work-As-Done (pp. 1-12). CRC Press.

Carayon, P., Schoofs Hundt, A., Karsh, B.-T., Gurses, A. P., Alvarado, C. J., Smith, M., & Flatley Brennan, P. (2006). Work system design for patient safety: The SEIPS model. Quality and Safety in Health Care, 15(Suppl 1), i50-i58. https://doi.org/10.1136/qshc.2005.015842

Crandall, B., Klein, G., & Hoffman, R. R. (2006). Working minds: A practitioner’s guide to cognitive task analysis. MIT Press.

Debono, D. S., Greenfield, D., Travaglia, J. F., Long, J. C., Black, D., Johnson, J., & Braithwaite, J. (2013). Nurses’ workarounds in acute healthcare settings: A scoping review. BMC Health Services Research, 13, Article 175. https://doi.org/10.1186/1472-6963-13-175

Ferneley, E. H., & Sobreperez, P. (2006). Resist, comply or workaround? An examination of different facets of user engagement with information systems. European Journal of Information Systems, 15(4), 345-356. https://doi.org/10.1057/palgrave.ejis.3000629

Halbesleben, J. R. B., Savage, G. T., Wakefield, D. S., & Wakefield, B. J. (2010). Rework and workarounds in nurse medication administration process: Implications for work processes and patient safety. Health Care Management Review, 35(2), 124-133. https://doi.org/10.1097/HMR.0b013e3181c9b561

Hollnagel, E. (2012). FRAM: The functional resonance analysis method: Modelling complex socio-technical systems. Ashgate.

Hollnagel, E., Wears, R. L., & Braithwaite, J. (2015). From Safety-I to Safety-II: A white paper. The Resilient Health Care Net. https://www.england.nhs.uk/signuptosafety/wp-content/uploads/sites/16/2015/10/safety-1-safety-2-whte-papr.pdf

Horwitz, L. I., Moin, T., Krumholz, H. M., Wang, L., & Bradley, E. H. (2009). Consequences of inadequate sign-out for patient care. Archives of Internal Medicine, 169(12), 1114-1120. https://doi.org/10.1001/archinternmed.2009.106

Nicolini, D., Mengis, J., & Swan, J. (2011). Understanding the role of objects in cross-disciplinary collaboration. Organization Science, 23(3), 612-629. https://doi.org/10.1287/orsc.1110.0664

Pronovost, P., Needham, D., Berenholtz, S., Sinopoli, D., Chu, H., Cosgrove, S., Sexton, B., Hyzy, R., Welsh, R., Roth, G., Bander, J., Kepros, J., & Goeschel, C. (2006). An intervention to decrease catheter-related bloodstream infections in the ICU. New England Journal of Medicine, 355(26), 2725-2732. https://doi.org/10.1056/NEJMoa061115

Reason, J. (1990). Human error. Cambridge University Press.

Tucker, A. L., & Edmondson, A. C. (2003). Why hospitals don’t learn from failures: Organizational and psychological dynamics that inhibit system change. California Management Review, 45(2), 55-72. https://doi.org/10.2307/41166165

van der Sijs, H., Aarts, J., Vulto, A., & Berg, M. (2006). Overriding of drug safety alerts in computerized physician order entry. Journal of the American Medical Informatics Association, 13(2), 138-147. https://doi.org/10.1197/jamia.M1809

Westbrook, J. I., Woods, A., Rob, M. I., Dunsmuir, W. T. M., & Day, R. O. (2010). Association of interruptions with an increased risk and severity of medication administration errors. Archives of Internal Medicine, 170(8), 683-690. https://doi.org/10.1001/archinternmed.2010.65