The Complaint May Be the Signal

The Complaint May Be the Signal

Using Operational Intelligence to Address Hospital Risk

Brian D. Baxter

It’s 3:00am on the 2nd floor of a medium-sized hospital in suburban America. A patient’s family complains that a nurse was rude, dismissive, and slow to respond. During the review, the hospital discovers that the same nurse made a medication-administration error earlier in the shift.

The initial facts appear straightforward. There is a complaint, an identifiable employee, a possible policy violation, and a preventable, adverse event. The familiar organizational response is equally straightforward: document the incident, counsel or discipline the nurse, require remedial training, and close the file. But what if someone were to look deeper?

They might find that the nurse in question has been working consecutive 12-hour shifts and volunteering for additional overtime. Her unit has been relying heavily on agency and travel nurses, some of whom earn substantially more per hour while working fewer shifts. Resentment has been growing among permanent staff. Break-room conversations have become divisive. Informal groups are forming, gossip is replacing productive communication, and the relationship between permanent and temporary staff is deteriorating.

On the floor, mandatory bedside reports are inconsistent at best. Medication-administration safeguards are sometimes bypassed because they are regarded as cumbersome during periods of high patient volume. Nurses hesitate to call certain physicians with important questions because previous calls were met with impatience or hostility. Supervisors know that shortcuts occur, but workload pressures and staffing needs make rigorous enforcement difficult. The written policies are there, but the normalization of deviance has allowed the actual work practices to drift away from them.

This is not a description of one documented hospital incident. It is a composite scenario built from conditions repeatedly identified in patient-safety research: fatigue, workload, interruptions, technology problems, communication breakdowns, cultural pressure, and normalized procedural deviations.

The Agency for Healthcare Research and Quality (AHRQ) reports that medication errors have been found to be three times more likely among nurses working shifts longer than 12.5 hours on more than two consecutive days.[1] Research summarized by AHRQ also identifies fatigue, heavy workload, interruptions, complacency, and insufficient medication knowledge as interconnected contributors to medication-administration errors.[2]

The complaint in this scenario is real enough. The mistake may be real. The nurse’s behavior still matters. But the event is also likely a red flag indicating operational risk that extends far beyond one employee. That is where operational intelligence begins.

Treating the Anomaly as the Entire Problem

Hospitals produce enormous amounts of information: censuses, incident reports, patient complaints, medication records, staffing schedules, overtime expenditures, employee-turnover data, security reports, workers’ compensation claims, patient-satisfaction surveys, and compliance findings. The problem is not necessarily a lack of information. The problem is that these data often remain siloed by department, purpose, and ownership.

Human resources sees the complaint. Nursing leadership sees the staffing shortage. Pharmacy sees the medication variance. Finance sees agency labor costs. Risk management sees the exposure. Security sees increasing aggression on the unit. Compliance sees deviations from policy. Senior leadership may receive pieces of all of it without receiving a coherent account of how the pieces interact.

When the organizational response consistently begins and ends with the employee closest to the event, the analysis is short-stroked. Leadership may identify who made the mistake without discovering the conditions that made the mistake more likely, allowed it to pass through existing safeguards, or caused similar practices to become normalized.

That response produces another operational consequence. Employees observe what happens to coworkers who report mistakes, question unsafe practices, or acknowledge procedural workarounds. If speaking candidly appears more likely to produce punishment than improvement, people become less willing to report near misses, emerging hazards, and uncomfortable truths. The organization does not eliminate the risk. It hides it.

Operational intelligence is the process of converting disconnected indicators into decision-ready knowledge. It asks what conditions surrounded the event, which safeguards were absent or ineffective, whether the incident is isolated or part of a pattern, and where actual practice has departed from written policy. It examines the incentives influencing behavior, the risks emerging across departmental boundaries, and the actions leadership can reasonably take.

This is the purpose of AIM: Analyze, Investigate, Mobilize.

Analyze: Understand the System Surrounding the Event

Analysis begins with the anomaly but does not stop with the employee involved.

In the medication-error scenario, the analysis would reconstruct the operational environment surrounding the nurse’s performance. That includes shift length, consecutive workdays, overtime, rest opportunities, patient load, patient acuity, interruptions, staffing composition, supervisory coverage, and recent exposure to threats, aggression, or workplace violence.

The medication-administration system must also be examined. Was the automated dispensing cabinet configured appropriately? Were alerts meaningful, or had staff become desensitized by frequent low-value warnings? Were workarounds common? Did the physical arrangement of equipment, medication, and patient information support accurate performance? Were policies current, understandable, and realistically executable under actual unit conditions? If so, were those policies known to the staff and consistently enforced by leadership?

Technology does not eliminate human fallibility. In some circumstances, it merely changes its form. AHRQ warns, for example, that overburdened clinicians may experience alert fatigue and fail to respond appropriately to automated notifications.[3]

Analysis should also determine whether the hospital has meaningful limits on consecutive hours, voluntary overtime, and required recovery time. A policy that exists but is routinely waived under staffing pressure is not functioning as an operational control. It has become an administrative artifact.

The Joint Commission’s Sentinel Event Policy reinforces the need for this broader view when a qualifying serious event occurs. The policy calls for a comprehensive systematic analysis of causal and contributory factors, followed by a corrective-action plan that is implemented and monitored for effectiveness.[4] Although not every complaint or medication variance constitutes a sentinel event, the same underlying principle remains valuable: understanding what happened is not the same as understanding why it happened.

The objective is not to construct an excuse for the nurse. It is to identify the conditions under which the behavior occurred so leadership can determine whether it is dealing with human error, an at-risk shortcut, reckless behavior, a system deficiency, or some combination.

Accountability remains necessary, but accountability without analysis is not honest accountability. If the response focuses exclusively on the person, leaders may correct one employee while preserving the conditions that will influence the next one. Honest accountability requires that expectations be clear, trained, known an crystalized within the organization. It also requires that the expectations not be beyond known limits of human performance.

Investigate: Discover How Work Is Actually Being Performed

Documents tell investigators how work is supposed to occur. Interviews and observations reveal what really happened.

The investigative phase should include confidential interviews across roles and levels, direct observation of relevant work processes, review of supervisory practices, and comparison of policy requirements with customary behavior. It might uncover that bedside reports are routinely skipped, medication scanning is inconsistently enforced, or calls to physicians are delayed because nurses anticipate an abrasive response. None of those practices are likely to appear in an official report. All of them can affect patient care.

The quality of an investigation depends upon whether employees believe they can speak honestly. Staff members who expect every disclosure to become a search for someone to punish will protect themselves, their coworkers, or their department. Interviews become guarded, reports become less detailed, and procedural deviations remain hidden until they contribute to a more serious outcome.

This is why a systems approach and honest accountability must be considered together. Employees should not be promised freedom from consequences regardless of their conduct. They should, however, be able to report human error, at-risk practices, unsafe conditions, and supervisory concerns without assuming that honesty itself will make them the next target. AHRQ’s just-culture framework recognizes this distinction: accountability remains, but the response is informed by the nature of the behavior and the system in which it occurred.[7]

The investigation would also examine staffing economics. What is the hospital spending on agency and travel nurses? How does that expenditure compare with the cost of improving compensation, scheduling, retention incentives, or working conditions for permanent staff? The answer cannot be assumed. It must be calculated using the hospital’s actual labor costs, turnover experience, vacancy periods, and mission-critical needs.

The travel nurse is not the problem merely because that nurse earns more. The operational concern is whether compensation differences, inconsistent expectations, and perceived inequities are contributing to diminished morale, resentment, poor teamwork, information withholding, or degraded communication. This is also where workplace violence enters the picture.

Hospital workplace violence is not limited to a physical attack by a patient or visitor. The broader risk environment can include threats, intimidation, bullying, harassment, hostile communication, fear of retaliation, and the normalization of abusive conduct. Workplace violence, burnout, and patient-safety culture are interconnected concerns, and reporting culture influences whether leaders see emerging problems before they become serious events.[5]

The Joint Commission’s workplace-violence requirements provide a useful framework for this work. For accredited hospitals, its National Performance Goals call for a workplace-violence prevention program developed by a multidisciplinary team and led by a designated individual. The program must include prevention and response policies, incident reporting and trend analysis, follow-up and support for affected employees, reporting to the governing body, and staff education and training.[6]

Those expectations cannot be satisfied meaningfully by maintaining a policy binder and conducting annual training. They depend upon reporting, analysis, leadership oversight, follow-up, and the conversion of information into action. In other words, they require an intelligence process.

A fatigued nurse who has been assaulted by a patient, criticized by a physician, disrespected by a coworkers, and dismissed by a supervisor does not enter the next patient interaction unaffected. Neither does a physician who is overloaded, a supervisor who is managing an understaffed unit, or a security officer repeatedly sent into volatile situations without reliable clinical support. While that does not make harmful conduct acceptable, it does make the surrounding conditions operationally relevant.

The Leadership Bandwidth Gap

One of the greatest obstacles to a meaningful AIM process is time.

Directors of nursing, house supervisors, charge nurses, risk managers, security leaders, and hospital administrators already manage staffing shortages, patient-flow demands, employee conflicts, clinical escalations, documentation requirements, regulatory obligations, and immediate patient-care concerns. They may recognize that a complaint or error deserves a deeper review while lacking the time to reconstruct the event, examine staffing and financial data, interview personnel, observe work practices, assess policy alignment, and coordinate findings across departments.

They also may not have been trained to conduct a multidisciplinary human-performance analysis. That is not an indictment of hospital leadership. Clinical and administrative leaders are trained for demanding roles requiring specialized expertise. A comprehensive AIM review, however, requires a different combination of operational analysis, investigative practice, human-performance knowledge, policy assessment, and implementation planning.

In many hospitals, first-line nursing leaders occupy an especially difficult dual role. They remain clinically or administratively engaged in the work while being expected to delegate, supervise, coach, correct, document, and evaluate the performance of others. Those are substantive leadership functions in any high-stakes profession. They cannot be treated as secondary duties to be performed only when patient volume permits.

When first-line supervision becomes primarily a matter of filling vacancies and resolving immediate disruptions, procedural drift can go unobserved. Informal practices replace approved procedures. Incivility becomes normalized. Known problems are handed from shift to shift. Senior leadership may remain unaware until the organization experiences a serious complaint, injury, medication event, employee departure, or act of violence.

First-line leaders also determine whether bad news travels upward. When their immediate response is to identify and correct the nearest employee, staff may learn to keep minor anomalies, workarounds, and near misses to themselves. Supervisors may then report that everything is functioning normally, not because the risks have been controlled, but because the signals have stopped reaching them.

A quiet unit is not necessarily a safe unit. Silence may indicate that employees trust the system and have few concerns. It may also indicate that they have learned there is little benefit, but considerable personal risk, in speaking up. This is the capacity gap AIM is intended to fill.

Performance-Critical brings the complete AIM process into the organization so hospital leaders can continue managing daily operations while a focused review examines the system surrounding the anomaly. The purpose is not to displace clinical judgment, nursing leadership, human resources, compliance, risk management, or security. It is to connect their information, examine the human-performance factors operating between them, and convert the findings into usable operational intelligence.

The same need exists in other high-risk industries. Leaders cannot suspend daily operations every time an anomaly requires deeper examination. Yet failing to examine it may allow an emerging risk to continue developing unnoticed. AIM creates the capacity to treat human performance as an operational priority without requiring leaders to lose sight of the work already in front of them.

Mobilize: Convert Findings Into Performance

A report alone does not change an organization. Operational intelligence becomes valuable when it is converted into action.

Mobilization begins with a prioritized corrective-action plan. Some measures may be immediate: addressing the individual medication event, protecting the affected patient, preserving evidence, satisfying reporting obligations, and determining the appropriate level of individual, honest accountability.

Other measures address the conditions revealed by the analysis and investigation. The hospital may need to revise limits on consecutive shifts and overtime, reassess the cost of permanent and temporary staffing, establish and enforce bedside-report expectations, or redesign medication workflows and equipment placement. It may need to reduce unnecessary alerts, clarify when physicians must be contacted, and establish enforceable behavioral expectations for those physicians.

Findings may also justify improving workplace-violence reporting and post-incident support, defining supervisory responsibility for correcting procedural drift, and providing first-line leaders with the time and authority necessary to manage employee performance. Leaders may need preparation in distinguishing human error, at-risk behavior, and reckless conduct. The organization may also need to monitor leading indicators rather than waiting for another adverse event to confirm that risk remains.

Policies are where organizational intentions are translated into expectations and SOPs are where those expectations are translated into performance. But publication alone is not implementation. Employees must understand the expectations, have a reasonable opportunity to comply, see supervisors model and reinforce it, and believe that it applies across professional rank and departmental boundaries. 

The familiar maxim says that culture eats strategy for breakfast. In operational environments, culture also eats policy for breakfast. If experienced staff routinely tell new employees, “That is what the policy says, but this is how we really do it,” the informal practice has become the operative system. Sending everyone through the same annual training again will rarely correct it. Leadership must identify why the workaround developed, determine whether the written process is workable, remove unnecessary obstacles, and then consistently enforce the expectation.

The same cultural conflict appears when a hospital’s policy tells employees to report hazards, near misses, abusive conduct, and procedural deviations, but the organization’s response teaches them that speaking up carries personal risk. The formal policy says, “Report it.” The informal culture says, “Keep your head down.” In daily operations, employees quickly learn which message carries more weight.

Mobilization therefore includes more than policy revision and classroom training. It requires organizational inculcation, supervisory participation, implementation support, reinforcement, and follow-up measurement. First-line supervisors are especially important because this is where organizational expectations encounter the pressures of actual work. They decide, moment by moment, what will be tolerated, reinforced, reported, corrected, or quietly ignored.

The Joint Commission’s model similarly extends beyond identifying a problem. Its Sentinel Event Policy expects organizations to implement corrective actions and monitor whether those actions are effective.[4] Closing the investigative file is not the same as mitigating the risk.

Beyond Name, Blame, Shame, Fire, or Retrain

A systems approach is sometimes misunderstood as a no-accountability approach. It is not. AHRQ describes a just culture as one that considers the type of behavior involved rather than responding solely according to the severity of the outcome.[7] Human error, at-risk behavior, reckless conduct and willful misconduct should not receive identical responses. Likewise, system deficiencies should not be concealed by disciplining the nearest employee.

The nurse in our scenario may need support, coaching, discipline, termination, or some combination of measures. That determination must follow the evidence. At the same time, disciplining one nurse will not correct chronic fatigue, dysfunctional communication, misaligned labor incentives, poorly designed medication processes, normalized shortcuts, or supervisory inattention.

The “name, blame, shame, fire, or retrain” cycle creates the appearance of action while potentially leaving the underlying risk intact. It also teaches the rest of the workforce an unintended lesson: when something goes wrong, keep your head down, say as little as possible, and avoid becoming associated with the event.

That lesson is costly. Complaints, near misses, informal concerns, unusual behavior, and procedural deviations are all potential signals. When employees stop reporting them, leaders lose the ability to recognize patterns and intervene early. The hazard remains present, but the operational intelligence needed to understand it disappears.

From One Complaint to Operational Intelligence

Hospital administrators and senior leaders do not need another dashboard filled with disconnected numbers. They need to know what those numbers, complaints, incidents, and behaviors mean when considered together.

The patient complaint may be a customer-service matter. It may reveal an individual performance problem. It may also be the first visible indication of fatigue, staffing instability, cultural division, procedural drift, workplace aggression, or failing supervisory controls.

AIM provides a disciplined way to find out. Analyze the event within the system that shaped it. Investigate the difference between written expectations and operational reality. Mobilize the organization to address system vulnerabilities, individual behavior, leadership practices, and cultural conditions together. Then remain engaged long enough to determine whether the intervention changed performance and whether employees trust the organization enough to continue reporting what they see.

Operational intelligence depends upon signals. Some arrive as complaints, errors, near misses, security incidents, policy deviations, or uncomfortable conversations. Others come from employees who are willing to say, “Something is not working.”

If the response is always to pursue the person closest to the event, the organization may close the case while leaving the risk in place. Worse, it may ensure that the next person remains silent.

Operational intelligence does not promise that hospitals can eliminate human error, workplace conflict, or violence. It gives leaders a better opportunity to recognize developing risk, make defensible decisions, satisfy organizational and accreditation expectations, and intervene before another complaint becomes a more consequential event. It also provides the analytical capacity to prioritize human performance without taking leaders away from the day-to-day responsibilities that cannot be placed on hold.

Sometimes the complaint is the problem. Sometimes it is the signal. Sometimes, the greatest warning is when the signals stop coming.

References

[1] Agency for Healthcare Research and Quality. “Nursing and Patient Safety.” Patient Safety Network. AHRQ: Nursing and Patient Safety

[2] Agency for Healthcare Research and Quality. “Nurses’ Perceived Causes of Medication Administration Errors: A Qualitative Systematic Review.” Patient Safety Network. AHRQ: Medication-Administration Errors

[3] Agency for Healthcare Research and Quality. “Technology as a Tool for Improving Patient Safety.” Patient Safety Network, March 29, 2023. AHRQ: Technology and Patient Safety

[4] The Joint Commission. “Sentinel Event Policy.” 2026. The Joint Commission: Sentinel Event Policy

[5] Kim, S., Lynn, M.R., Baernholdt, M., Kitzmiller, R., and Jones, C.B. “How Does Workplace Violence-Reporting Culture Affect Workplace Violence, Nurse Burnout, and Patient Safety?” Journal of Nursing Care Quality 38, no. 1 (2023): 11–18. AHRQ summary

[6] The Joint Commission. “National Performance Goals Effective January 2026 for the Hospital Accreditation Program.” NPG.02.04.01, Workplace Violence Prevention Program. The Joint Commission: National Performance Goals

[7] Agency for Healthcare Research and Quality. “Culture of Safety.” Patient Safety Network. AHRQ: Culture of SafetyThe