The 2016 Medicare emergency rule named cyberattacks as a hazard and left hospitals to rehearse the outage

The emergency preparedness rule published September 16, 2016, covered 17 provider and supplier types in Medicare and Medicaid. It named cyberattacks among possible hazards but declined to write specific cyber security requirements, so training, exercises and records had to carry that risk.

Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.

Two-minute summaryThe dek and the first sentence of each section, quoted from the article

The emergency preparedness rule published September 16, 2016, covered 17 provider and supplier types in Medicare and Medicaid. It named cyberattacks among possible hazards but declined to write specific cyber security requirements, so training, exercises and records had to carry that risk.

  1. The week The rule reached widely.
  2. The risk assessment: what the all-hazards assessment had to cover A second place was the comment response described in the week section.
  3. The people: training, staffing, volunteers, tracking and communication The phrase “consistent with their expected role” does work.
  4. The records: medical documentation and the release of patient information The hospital’s policies had to provide a system of medical documentation that preserves patient information, protects its confidentiality, and secures and maintains the availability of records. [2] That single clause is the closest the rule came to an IT requirement.
  5. The exercises: drills, documentation and power The rule defined the tabletop option.
  6. How the rule lands on different settings For a system that shares one EHR and one network, a cyber event is a system-level hazard.
  7. Objections, in order There is a real division of labor.
  8. With hindsight CMS revisited the rule in 2019.

The week

On September 16, 2016, the Federal Register published a final rule from the Centers for Medicare & Medicaid Services on emergency preparedness. [2] It set emergency preparedness requirements for providers and suppliers that take part in Medicare and Medicaid. [2] The agency had announced the rule on September 8, 2016. [1]

The rule took effect on November 15, 2016. [2] Facilities had until November 15, 2017, to implement it. [2] The agency’s release described the timing as effective 60 days after publication, with full implementation required one year after the effective date. [1]

The rule reached widely. In its responses to comments, the agency said its approach addressed the differences between the 17 provider and supplier types the regulations covered. [2] For hospitals, the requirements sat in a new condition of participation at 42 CFR 482.15. [2]

The release framed the gap the rule was filling. It said a review of existing Medicare emergency preparedness rules found them not comprehensive enough. [1] The release said they did not address communication to coordinate with other systems of care, contingency planning, or training of personnel. [1]

The release also pointed to recent events. It said a number of disasters over the past several years, most recently in Louisiana, had put beneficiaries and the public at risk. [1] The rule’s own background listed earlier events, from the September 11, 2001 attacks to Hurricane Sandy in 2012. [2]

The rule organized the requirements into four elements. The agency listed them as risk assessment and emergency planning, policies and procedures, a communication plan, and a training and testing program. [1][2] The release said the standards were adjusted to the characteristics of each provider and supplier type. [1]

The first element carried the hazard list. The rule required an all-hazards risk assessment before a provider built its emergency plan. [2] The rule said that approach considers the hazards most likely in a provider’s area. [2] These may include care-related emergencies, equipment and power failures, and interruptions in communications, including cyber-attacks. [2]

Cyber security came up again in comments. A commenter asked that communications planning include interoperability, redundancy and cyber security provisions. [2] The agency said it agreed hospitals should consider security, interoperability and redundancy in their plans. [2] It then said it had not included specific requirements on cyber security and redundancy, and encouraged providers to assess whether they could benefit from such plans. [2]

The rule declined another technology mandate too. Some commenters asked whether the required system of medical documentation had to be electronic. [2] The agency said it was not requiring electronic health records. [2] It said it expected providers to preserve and protect patient records and keep them secure for continuity of care. [2]

The agency also changed some proposals in response to comments. Its release said it removed a requirement for additional hours of generator testing. [1] It said it gave providers flexibility to choose the type of exercise for their second annual test. [1] And it allowed a separately certified hospital or other provider in a health system to join the system’s unified emergency program. [1]

The rule set specific duties for some settings and not others. The release said hospitals, critical access hospitals and long-term care providers had to install and maintain emergency and standby power systems based on their emergency plan. [1] It said outpatient providers such as ambulatory surgical centers and end-stage renal disease facilities would not need policies for subsistence needs. [1]

The headline on September 8 was consistency: one set of preparedness rules across Medicare and Medicaid settings. [1] For anyone running a hospital’s technology, the more telling line came in the comment responses. CMS agreed cyber security mattered, then declined to write specific requirements for it. [2]

Our argument is that the 2016 rule put cyber risk inside a program built for people and operations. The rule named cyberattacks only as an example hazard. [2] What it enforced was the machinery around a plan: training, exercises, contact lists, staff tracking and documentation. [2] In our reading, that left each hospital to decide whether its drills would include losing its own systems.

We take the rule in four parts:

  1. The risk assessment: what the all-hazards assessment had to cover, and where cyber appeared.
  2. The people: training, staffing, volunteers, tracking and the communication plan.
  3. The records: the system of medical documentation and the release of patient information.
  4. The exercises: the required drills, their documentation and the power systems behind them.

The first part decides what the plan is for. The second covers who carries it out, the third what information they need, and the fourth how the hospital proves it works. Together they cover the parts of the hospital condition at 42 CFR 482.15, as written in 2016, that an outage tests. [2] After them we look at how the rule lands on different settings, take the strongest objections in order, and end with checks an emergency manager and a CIO could run together.

The risk assessment: what the all-hazards assessment had to cover

The text of the requirement

The 2016 hospital condition required a comprehensive emergency preparedness program using an all-hazards approach. [2] The plan had to be based on a documented, facility-based and community-based risk assessment. [2] It had to include strategies for the emergency events that assessment identified. [2]

The plan also had to address the patient population, including people at risk. [2] It had to cover the services the hospital could provide in an emergency and continuity of operations, including delegations of authority and succession plans. [2] In 2016, the plan and the assessment had to be reviewed and updated at least annually. [2]

Where cyber appeared

The word “cyber” appears in the rule’s summary of the all-hazards approach. [2] Cyber-attacks were listed as one kind of interruption in communications, among other hazards such as power failures and loss of part of a facility. [2] The regulation text for hospitals does not mention cyber at all. [2]

A second place was the comment response described in the week section. CMS said hospitals should consider security, interoperability and redundancy, but that it had not included specific requirements on cyber security. [2] It encouraged facilities to assess whether they could benefit from such plans. [2]

A third mention sits in the rule’s statement of benefits. There the agency described Presidential Policy Directive 8 as aimed at the threats that pose the greatest risk to the nation, including acts of terrorism, cyber-attacks, pandemics and catastrophic natural disasters. [2] That passage explains the rule’s purpose; it sets no duty.

In our reading, that response made cyber a matter for the risk assessment. A hospital whose assessment named a cyberattack as a likely hazard would have to plan strategies for it. A hospital whose assessment did not could still meet the condition on paper.

Why CMS kept the rule general

CMS explained its general approach in the same responses. It said rules that were too specific could become outdated as technology and the nature of threats change. [2] It said rules that were too broad could be ineffective. [2] It chose four main components tied to the National Preparedness System, tailored by provider type. [2]

That reasoning has force. A 2016 rule that named specific security controls might have aged poorly. But the choice had a cost. It left the definition of a cyber emergency to each facility’s planners, whatever their background.

What a cyber hazard means for a hospital

A cyberattack that disables clinical systems is a strange emergency in the rule’s terms. The building stands. Power may work. Staff are present. What is missing is information: orders, results, schedules, medication records and the phones or pagers that depend on the network.

The rule’s hazard list already contained the pieces. It named equipment failures and interruptions in communications. [2] A ransomware event combines both. In our reading, a risk assessment that treats it as a communications interruption alone would understate it.

The community side of the assessment

The assessment had to be community-based as well as facility-based. [2] The plan needed a process for working with local, tribal, regional, state and federal emergency officials. [2] In 2016, that included documenting the hospital’s efforts to contact those officials and its participation in collaborative planning. [2]

For a cyber event, the community view matters in a specific way. If one hospital diverts ambulances, neighbors absorb the load. An attack on a shared vendor could reach several hospitals at once. A community-based assessment is the natural place to record that dependency.

Dependencies outside the building

An all-hazards assessment that stops at the hospital’s walls misses a class of cyber hazard. A hospital may rely on outside services for claims, pharmacy benefits, lab results, imaging reads or its hosted record system. An attack on any of them can stop work inside a hospital that was never breached.

The 2016 rule’s community-based assessment gave a hook for that analysis. [2] The communication plan already required contacts for entities providing services under arrangement. [2] In our reading, a hospital could list its critical technology suppliers in the same way, with what each supports and what the fallback is when it fails.

HHS later described this risk in its own words. Its strategy paper said cyber incidents had caused extended care disruptions from multi-week outages and patient diversion. [5] The paper did not tie those outages to the 2016 rule. It does show the kind of event the hazard list needed to cover.

Who owns the assessment

The rule did not name an owner. Emergency planning and technology risk can sit with different leaders, under different rules. The HIPAA Security Rule already required a contingency plan for emergencies that damage systems holding electronic protected health information. [8]

That split matters. One program asks how the hospital keeps caring for patients. The other asks how it protects and restores data. In our reading, the 2016 rule’s silence on cyber left those two programs to meet only if someone made them meet.

The people: training, staffing, volunteers, tracking and communication

The training program

The 2016 condition required initial training in emergency policies for all new and existing staff, people providing services under arrangement, and volunteers, consistent with their roles. [2] It required training at least annually, documentation of training, and a demonstration of staff knowledge of emergency procedures. [2]

The phrase “consistent with their expected role” does work. A nurse’s role in an outage differs from a registration clerk’s or a pharmacist’s. Training that covers evacuation routes but not paper ordering would leave clinical staff unready for a system outage.

The demonstration requirement asks for more than attendance. A hospital has to show staff know the procedures. [2] For a downtime scenario, that might mean a nurse completing a paper medication record in a drill, observed and recorded.

Staffing strategies

The policies and procedures had to address the use of volunteers in an emergency and other emergency staffing strategies. [2] They had to cover the integration of state and federally designated health professionals to address surge needs. [2]

Surge staffing can mean more hands for more patients. A cyber outage creates a different surge. Patient numbers may not change, but each task takes longer on paper. Runners carry results, clerks reconcile records and pharmacists verify orders by hand. In our reading, staffing strategies that ignore that slowdown would miss the main labor effect of an outage.

Tracking staff and patients

The policies had to include a system to track the location of on-duty staff and sheltered patients during an emergency. [2] If patients or staff were relocated, the hospital had to document the name and location of the receiving facility. [2]

CMS addressed the tracking requirement as one of its general comment topics. [2] For technology leaders, the point is dependency. A tracking system that runs on the network that just failed tracks nothing. A paper fallback, rehearsed, meets the requirement in the conditions where it matters.

The communication plan

The communication plan had to list names and contact information for staff, entities providing services under arrangement, patients’ physicians, other hospitals and volunteers. [2] It had to list contacts for emergency preparedness staff at all levels of government and other sources of assistance. [2]

It also had to include primary and alternate means of communicating with staff and with emergency management agencies. [2] In 2016, the plan had to be reviewed and updated at least annually. [2]

The alternate means is the cyber-relevant part. If the primary means is email, a hosted phone system or a paging platform on the hospital network, an attack may take it down. The rule did not say what the alternate must be. In our reading, an alternate that shares a network or identity system with the primary is not much of an alternate.

Information for the authorities

The communication plan also had to include a means of giving the authority having jurisdiction, the Incident Command Center or a designee information about the hospital’s occupancy, needs and ability to help. [2] In a regional disaster, that report helps officials place patients.

In a cyber outage, the same report can be hard to produce. Bed counts, staffing levels and supply data may all come from systems that are down. In our reading, the plan should say how the hospital would assemble that report by hand, and who would send it.

Where the contact list lives

A contact list is data. If it lives only in an HR system or a shared drive, an attack can make it unreachable. The rule required the list, not its location. [2] Printed copies, offline copies and copies held by a partner hospital are all design choices a hospital could make within the rule.

People who work under arrangement

The training duty reached beyond employees. It covered individuals providing services under arrangement and volunteers, consistent with their roles. [2] The communication plan had to list entities providing services under arrangement. [2]

In a hospital, contracted staff can include agency nurses, physician groups and outsourced services such as transport or environmental services. Some hospitals also contract out technology operations. In our reading, a managed service provider that runs the network or the help desk plays a role in an outage, and the training duty gives a reason to include its staff in drills.

Training records as evidence

The rule required the hospital to maintain documentation of training. [2] That record answers a surveyor’s first question: who was trained, on what, and when. For a downtime scenario, it also shows which units have practiced paper workflows and which have not.

A training record that lists only attendance says little about readiness. One that records the scenario, the role and the demonstration of knowledge gives the hospital a way to find gaps. [2] The rule required the demonstration; the format was the hospital’s choice.

Delegations and succession

The plan had to address continuity of operations, including delegations of authority and succession plans. [2] In a cyber event, the people who hold authority may be present but cut off from the tools they use to exercise it. Approving purchases, diverting ambulances and calling in staff all depend on systems.

A delegation written for a hurricane assumes absent leaders. A delegation written for an outage assumes present leaders without tools. In our reading, a plan that covers both would say who decides, how decisions are recorded on paper, and who restores the record afterward.

The records: medical documentation and the release of patient information

What the rule required

The hospital’s policies had to provide a system of medical documentation that preserves patient information, protects its confidentiality, and secures and maintains the availability of records. [2] That single clause is the closest the rule came to an IT requirement.

CMS explained what it did not mean. It said it was not requiring electronic health records. [2] It said facilities were at varying stages of EHR adoption. [2] It expected facilities to provide a means to preserve and protect patient records and keep them secure, for continuity of care. [2]

Waivers and alternate care sites

The policies had to address the hospital’s role under a waiver declared by the Secretary under section 1135 of the Social Security Act, for care at an alternate care site identified by emergency officials. [2] That provision assumes a declared emergency.

A cyberattack on one hospital may not lead to any declaration. In our reading, a plan that leans on waiver flexibility for an outage could find none available. The plan would need to work under normal rules, with paper records and diversion arranged through ordinary agreements with other providers. [2]

Why availability is the hard word

Preserving records protects against loss. Protecting confidentiality guards against exposure. Maintaining availability is different: the records must be usable when needed, during the emergency itself. [2]

A cyberattack attacks availability directly. The records may be intact and private, yet locked. In our reading, a hospital that reads the clause only as a backup duty would miss the word that matters in an outage. A backup that takes days to restore does not keep records available on the first night.

What commenters said about paper

The comment record shows hospitals already thought about this in physical terms. Some commenters said that if power were lost, they could not copy records or reach patient records on computers. [2] Some said they used pre-printed paper forms attached to patients during an evacuation. [2]

Those comments describe downtime procedures. A hospital that maintains paper forms for evacuation has a starting point for a cyber outage. The forms, the habit of using them and the people who know where they are stored all carry over. The difference is duration. An evacuation hands a patient off. An outage may keep the patient in place for days on paper.

Sharing information with other providers

The communication plan had to include a method for sharing information and medical documentation with other providers to maintain continuity of care. [2] It had to include a means, in an evacuation, to release patient information as permitted under the HIPAA Privacy Rule’s provisions on disclosures to family and others. [2]

CMS said it expected plans to address how a provider would release patient information in an evacuation, as the HIPAA Privacy Rule permits. [2] It suggested plans say what information is releasable and who may release it. [2] Those are governance questions, and they apply in an outage too, when staff may be tempted to send records by personal email or phone.

Confidentiality on paper

The records clause asked for confidentiality as well as availability. [2] Paper records created during an outage carry patient information outside the controls of the electronic system. They sit at nursing stations, travel with runners and pile up for later entry.

In our reading, a downtime procedure needs a privacy step. It would say where paper records are kept, who may see them, how they move between units and how they are stored or destroyed once entered. Without that step, a hospital can keep records available while losing track of who saw them.

The HIPAA Security Rule required covered entities and business associates to have a contingency plan for emergencies that damage systems containing electronic protected health information. [8] That plan required a data backup plan, a disaster recovery plan and an emergency mode operation plan. [8] Testing and revision of contingency plans was an addressable specification, as was a criticality analysis of applications and data. [8]

In our reading, the two rules meet at the medical documentation clause. The CMS condition asks that records stay available for care. [2] The HIPAA plan asks how systems and data are restored and how critical processes continue in emergency mode. [8] A hospital that runs them as separate programs risks plans that assume each other.

Restoration and the backlog

The HIPAA contingency plan required a disaster recovery plan to restore any loss of data, and a data backup plan to keep retrievable exact copies of electronic protected health information. [8] Those are required implementation specifications. [8] They address the end of an outage.

The CMS condition addresses the middle. While systems are down, care continues on paper or in other tools. [2] When systems return, that care must be entered or linked to the electronic record. In our reading, the backlog is a patient safety risk of its own. Orders given on paper and not entered can be repeated or missed.

A plan that covers restoration without the backlog leaves a gap between the two rules. A hospital could close it by naming who enters the paper record, in what order, and how clinicians are told what has not yet been entered.

What the record system needs during a drill

A drill that tests documentation would look at several things. It would check whether clinicians can see current medication lists without the main system. It would check whether they can record care on paper in a form that can be entered later. And it would name who reconciles the paper record when systems return, and how the gap is documented.

None of those checks appear in the rule. All of them follow from its requirement to maintain the availability of records. [2] The rule left the method to the hospital.

The exercises: drills, documentation and power

What the 2016 rule required

The 2016 hospital condition required exercises to test the emergency plan at least annually. [2] A hospital had to take part in a full-scale exercise that is community-based, or, when that is not accessible, an individual facility-based one. [2] It also had to run an additional exercise, which could be a second full-scale exercise or a tabletop exercise. [2]

The rule defined the tabletop option. It described a facilitated group discussion using a narrated, clinically relevant emergency scenario and a set of problem statements, directed messages or prepared questions designed to challenge the plan. [2]

A hospital that activated its plan for a real emergency was exempt from the full-scale exercise for 1 year following the onset of the event. [2] The hospital had to analyze its response to drills, tabletop exercises and real events, keep documentation, and revise the plan as needed. [2]

When a real event counts

The exemption for a real event depended on activation of the emergency plan. [2] The rule spoke of an actual natural or man-made emergency. [2] In our reading, a cyberattack that forced the hospital to activate its plan would be a man-made emergency, and its after-action analysis would serve the same purpose as a drill’s.

That reading has a practical edge. A hospital that does not activate its plan during an outage, and handles it as an IT incident, gets no exemption and also loses the learning the documentation duty was meant to capture. [2]

The choice of scenario

The rule did not choose the scenario. That choice belonged to the hospital, guided by its risk assessment. [2] A hospital that rehearsed a hurricane every year could meet the rule without rehearsing an outage.

In our reading, the tabletop option was a low-cost way to close the gap. A facilitated discussion of a ransomware scenario costs a morning. It would expose the dependencies the plan assumed away: the paging system, the badge readers, the pharmacy cabinets, the lab interface.

Exercising with the community

The full-scale exercise was meant to be community-based where possible. [2] CMS encouraged smaller hospitals to engage with healthcare coalitions in their area for help meeting the requirements. [2] It said it was not mandating that smaller facilities confer with local emergency response networks. [2]

A community exercise built around a cyber scenario would test things a single hospital cannot. It would show whether neighbors can absorb diverted patients, whether emergency medical services can reroute, and whether a public health agency can share information when one hospital’s systems are dark.

Documentation as evidence

The requirement to analyze and document every drill and real event matters for surveyors. [2] It turns an exercise into evidence. A surveyor can ask what the last drill found and what changed in the plan afterward.

For technology leaders, the after-action record is a useful artifact the rule produced. A note that the downtime forms were out of date, or that the alternate phone tree failed, is a finding with an owner. Without the documentation duty, those findings stay in someone’s memory.

Generator testing and the power analogy

The 2016 hospital condition required emergency and standby power systems based on the plan. [2] Generators had to be located according to NFPA codes when a building was new or renovated. [2] The hospital had to follow the inspection, testing and maintenance requirements in the Health Care Facilities Code, NFPA 110 and the Life Safety Code. [2]

CMS dropped a proposal for additional generator testing. It said, after reviewing comments and reports on Hurricanes Sandy and Katrina, that there were not sufficient data to assume more testing would ensure generators survive all disasters. [2] It expected facilities to keep testing under NFPA codes. [2]

The comparison is useful. Power had a detailed testing regime by reference to codes. Information systems had none in this rule. In our reading, the rule treated electricity as infrastructure and computing as a facility’s own business.

Fuel and duration

Hospitals with onsite fuel for emergency generators had to plan how to keep emergency power running during the emergency, unless they evacuated. [2] That is a duration question: how long the hospital can operate on its own resources.

A cyber outage raises the same question about information. How long can the hospital run on paper, with the staff it has, before care quality falls? The rule asked it for fuel. It did not ask it for data.

How the rule lands on different settings

A multi-hospital health system

The rule let a hospital in a system with several separately certified facilities join a unified, integrated emergency program. [2] Each facility had to take part in developing it, and the program had to account for each one’s circumstances, patients and services. [2] It still needed a documented facility-based risk assessment for each facility. [2]

For a system that shares one EHR and one network, a cyber event is a system-level hazard. A unified program is a natural place to plan for it. The facility-level assessment still matters, because a small hospital in the system may have fewer staff to run on paper.

A shared platform also changes the exercise design. A drill at one hospital tests that hospital’s downtime procedures. A system-wide outage tests whether every facility can run on paper at once, with the same central support teams stretched across all of them. The unified program had to show each facility could actively use the program. [2] In our reading, a system-wide outage drill is the clearest way to show that.

A hospital on a hosted record system

A hospital that runs its record system through an outside host depends on that host’s recovery as well as its own. The 2016 rule did not address hosting. [2] The HIPAA contingency plan still applied to the covered entity; 45 CFR 164.308 places its safeguards on covered entities and business associates alike. [8] In our reading, a host handling electronic health information on the hospital’s behalf would usually be a business associate.

In our reading, the emergency plan for such a hospital needs two scenarios: its own network down, and the host down. The fallback may differ. A local read-only copy of key data helps in the first case and may also help in the second, if it is kept current.

A critical access hospital

The CMS release said critical access hospitals, like hospitals, had to install and maintain emergency and standby power systems based on their plans. [1] A rural critical access hospital may have few nearby hospitals to receive patients. The rule required arrangements with other providers to receive patients if operations were limited. [2]

For a cyber event, distance raises the stakes. Diversion may mean long transfers. A small team may carry the paper workload alone. The risk assessment for such a hospital has reason to treat an outage as a high-impact hazard.

Ambulatory and dialysis providers

The release said outpatient providers and suppliers, such as ambulatory surgical centers and end-stage renal disease facilities, would not need policies for subsistence needs. [1] Their programs still followed the four-element structure. [1]

An ambulatory surgery center can cancel cases during an outage. A dialysis facility has less room, because patients need treatment on a schedule. In our reading, the risk assessment for a dialysis provider would look at how treatment records and machine settings are kept when systems fail.

Long-term care facilities

The release said long-term care facilities had to install and maintain emergency and standby power systems based on their plans. [1] Residents may stay for long periods, and their records support daily care. An outage there can affect medication administration across the facility.

Home health agencies

Home health agencies were among the provider types the rule covered. [2] Their staff work in patients’ homes and may depend on mobile devices and scheduling systems. An outage can leave clinicians without visit schedules or patient addresses. A contact list and visit schedule held offline is the basic fallback.

Objections, in order

1. Cyber security belongs in HIPAA, not in conditions of participation

There is a real division of labor. The HIPAA Security Rule already required a contingency plan for systems holding electronic health information. [8] The CMS rule did not need to duplicate it. But the HIPAA plan covers data and systems. The CMS condition covers patient care during the emergency. [2][8] An outage tests both at once.

2. Naming specific cyber controls would have dated quickly

CMS made this argument itself. [2] It is a fair one. Yet the objection is to specific controls, not to scenarios. A rule could have required hospitals to consider loss of information systems without naming a single control. The 2016 rule left even that to the risk assessment. [2]

3. The rule already listed cyberattacks, so hospitals were on notice

The rule did list cyber-attacks as an example. [2] An example in a preamble is weaker than a requirement in regulation text. In our reading, a surveyor could not cite a hospital for omitting a cyber scenario if its risk assessment did not identify one.

4. Hospitals already had downtime procedures

Some did, as the comments on paper forms suggest. [2] The rule’s value was to tie those procedures to training, exercises and documentation that surveyors could check. A downtime procedure nobody drills is a document, not a capability.

5. More requirements would have strained small facilities

Commenters raised cost and logistics, and CMS acknowledged those challenges when it dropped extra generator testing. [2] The burden concern is real for small and rural facilities. A tabletop exercise on an outage scenario, though, fits inside the second exercise the rule already required. [2]

6. Surveyors are not equipped to judge cyber readiness

Surveyors check conditions of participation, not network architecture. They are not security assessors, and the rule did not make them so. That is true, and the rule did not ask them to assess security controls. [2] What surveyors can check is whether the risk assessment names a hazard, whether staff were trained and whether exercises were documented. [2] Those checks work for an outage scenario as well as for a flood.

Checks an emergency manager and a CIO could run together

The risk assessment

  • Confirm the all-hazards assessment names loss of clinical and business systems as a hazard, with likelihood and impact.
  • Record which systems the plan assumes will work: paging, phones, badge access, pharmacy cabinets, the EHR, lab and imaging interfaces.
  • Note shared vendors whose failure could reach several facilities at once.

The people

  • Check that training for each role covers working without systems, consistent with the role.
  • Confirm the contact list and alternate communication means are reachable when the network is down.
  • Write the delegation for leaders who are present but without their tools.

The records

  • Test whether clinicians can see current medication lists and allergies during an outage.
  • Confirm downtime forms are current and stocked where care happens.
  • Name who reconciles paper records when systems return, and how the gap is documented.

The suppliers

  • List each outside service whose failure would stop clinical or business work, with the fallback for each.
  • Confirm which suppliers’ staff take part in drills, and how they are reached during an outage.
  • Check that a hosted system’s recovery commitments match what the emergency plan assumes.

The exercises

  • Run at least one tabletop exercise with an outage scenario and record the findings.
  • Track each finding to an owner and a plan change, and keep the record for surveyors.
  • Line up the HIPAA contingency plan test with the emergency exercise, so data restoration and patient care are tested together.

With hindsight

CMS revisited the rule in 2019. On September 30, 2019, it published a final rule on program efficiency and burden reduction that changed several emergency preparedness requirements. [4] It said applicable providers would review their emergency programs every two years, except long-term care facilities, which would still review annually. [4]

The 2019 rule also changed training. It required training every two years after initial training, except for long-term care facilities, and added training when the plan is significantly updated. [4] It expanded the types of exercises inpatient providers could use. [4] It also removed the requirement to document efforts to contact emergency preparedness officials. [4]

The current hospital text reflects those changes. As checked on November 16, 2026, 42 CFR 482.15 required review of the emergency plan, policies, communication plan and training program at least every 2 years. [3] It required exercises at least twice per year. [3] The eCFR source note listed amendments of November 16, 2016, and September 30, 2019. [3]

The current text still contains the 2016 records clause. A hospital must have a system of medical documentation that preserves patient information, protects confidentiality, and secures and maintains the availability of records. [3] As checked on November 16, 2026, the section contained no requirement that mentions cyber security. [3]

HHS later signaled interest in more. A strategy paper on health care cybersecurity, captured by the Internet Archive on January 9, 2024, said cyber incidents had led to multi-week outages and patient diversion. [5] It said HHS expected to seek comment on proposals based on sector cybersecurity performance goals. [5] One listed action was that CMS would propose new cybersecurity requirements for hospitals through Medicare and Medicaid. [5]

The same paper said the HHS Office for Civil Rights would begin an update to the HIPAA Security Rule in spring 2024. [5] That update arrived as a proposed rule published January 6, 2025. [6] Its preamble cited reporting on disruptions from the 2024 Change Healthcare cyberattack. [6]

The proposal reached the outage problem directly. It would require written procedures to restore critical relevant electronic information systems and data within 72 hours of a loss. [6] It would add a required criticality analysis of systems and assets. [6] It would require review and testing of contingency plans at least once every 12 months, with the results documented. [6] Comments were due March 7, 2025. [6]

As checked on November 16, 2026, the proposal was not final. The 2026 Unified Agenda listed it as a long-term action and projected final action for July 2027. [7] The current Security Rule still treated testing and revision of contingency plans as addressable. [8]

What held up from 2016 is the structure. The four elements, the training and exercise duties and the records clause remain in the hospital condition. [3] In our assessment, what looks weaker now is the choice to leave cyber risk to each facility’s own assessment. HHS’s own paper later described multi-week outages at hospitals and health systems, and listed new CMS cybersecurity requirements for hospitals among planned proposals. [5]

What a hospital leader can take from the record is a design choice. In our reading, the CMS condition and the HIPAA contingency plan already cover the main pieces an outage needs. [3][8] The drills, the paper forms, the contact lists and the restoration plan only work as a set if someone has rehearsed them together.

Spotted an error? Tell us and we will check it and correct the article where needed.

  • emergency preparedness
  • conditions of participation
  • CMS
  • exercises
  • downtime
  • HIPAA