Surgery-Day Downtime Plan for Oral Surgery Practices: Server, Internet, Power and Ransomware During Sedation Cases

It is 8:40 on a Tuesday and the second IV sedation case of the day is under way. WinOMS freezes on the front-desk PC, and a minute later the recovery-bay workstation shows the same spinning cursor. The vitals monitor in the suite still shows numbers, but nobody is sure whether the anesthesia record is still receiving them. A general dentist is on hold about a CBCT referral. Which of these is a five-minute problem, and which one closes the office for the afternoon?

This guide gives you that plan in pieces you can print: a seven-row scenario table, the contents of a downtime box for each surgical suite, a 30-minute pre-flight for surgery mornings, a restore-point worksheet and twelve questions for your IT provider and software host. It is written for the owner, administrator or office manager, and it covers the technology around the patient, not the clinical care. For the wider picture of dental IT support for oral surgery practices, and how downtime fits with the rest of the office, see the oral surgery practice IT guide.

This is the surgery-day overlay. For a full ransomware recovery playbook, see the dental ransomware recovery playbook; for internet troubleshooting, see what to check when the office internet goes down.

What is a downtime plan, and why is an oral surgery office different?

In IT terms, downtime means a system outage: a stretch when the practice software, the server, the internet, power or a device link is unavailable. Downtime procedures are the written steps staff follow until it comes back: paper documentation, communication that does not depend on the failed system, and a defined route back into the record.

The phrase comes from hospitals, and their guidance is a useful starting point. The federal SAFER guide on contingency planning asks that "paper forms are available to replace key EHR functions during downtimes", that "staff are trained and tested on downtime and recovery procedures", and that "the communication strategy for downtime and recovery periods is independent of the computing infrastructure that supports the EHR." Hospitals often turn that into downtime kits kept on each unit. In an oral surgery office, the same idea becomes a box in each surgical suite and a one-page table at the front desk.

Your software vendor sets the stakes in its own words. Carestream's technician guide for WinOMS calls it "a critical systems application that must be running every day with as little down time as possible." The same guide says "the local technician is fully responsible for the configuration, installation, and maintenance of the client's computer network," and that "WinOMS support does not train offices on configuring the backup system or restoring a database." In other words, the plan belongs to the practice and its IT provider, not to the software vendor.

Someone has to own it on the day. Name a downtime lead for every surgery day (often the office manager or the lead surgical assistant), and a second person for days the lead is out. The lead decides when the office switches to paper, makes the first call and decides when the office goes back to the software. If it helps to see where downtime sits among the other things that can stop a dental office, our map of the technology failures that can shut a dental practice lays them side by side.

What the regulation's text says

The HIPAA Security Rule includes a contingency plan standard at 45 CFR 164.308(a)(7). Its text reads: "Establish (and implement as needed) policies and procedures for responding to an emergency or other occurrence (for example, fire, vandalism, system failure, and natural disaster) that damages systems that contain electronic protected health information." It lists a data backup plan, a disaster recovery plan and an emergency mode operation plan as "Required", and testing and revision procedures and an applications and data criticality analysis as "Addressable".

This is the regulation's wording. How it applies to your practice is a question for your compliance adviser.

What makes a sedation day different

What keeps working when each system fails? The surgery-day scenario table

Most failures leave something running. An on-premises server keeps the schedule and charts available inside the office when the internet drops; a hosted session does not. The table below shows, for seven common failures, what usually keeps working, what stops, the paper fallback, who to call first and the restore point to ask for.

Print it, fill in the "call first" names and numbers for your office, and keep a copy in each downtime box. "Usually" matters: your own setup decides the details, so walk through each row with your IT provider once.

Vendor details in this table come from Carestream's WinOMS 10.1.x and 10.1.x Hosted system requirements and its technician guide, checked 26 October 2026.

Three rows need a closer look:

• • Row 1 vs row 2. With the server in the office, an internet outage is an inconvenience: the schedule, charts and anesthesia module keep working on the office network, because Carestream says the anesthesia module "must be locally installed through LAN or VPN connection." A server outage is the bigger problem, and it is where the paper record takes over.
• • Row 3. With a hosted or cloud system, the internet line is the system. Whether vitals keep recording during a session drop is not something the vendor's published requirements answer, so ask your host. Sensei's help pages describe manual vitals entry and say a monitor shows as disconnected when its connected PC is off or its service stops; they do not describe what happens to a recording that is in progress when the link drops.
• • Row 6. CISA's #StopRansomware Guide says to "determine which systems were impacted, and immediately isolate them", and to power devices down only if you cannot disconnect them from the network, because powering off loses evidence held in memory. Unplug network cables or let IT cut the switch; do not wipe or reinstall anything.

Hosted, cloud or on-prem: how your software choice changes the plan

The deployment you run decides which rows matter most. With WinOMS on a server in the office, rows 1 and 4 are the big ones: the server is the single point that stops the day, and a power cut that crashes it can damage the database. With WinOMS Hosted or Sensei Cloud, row 3 moves to the top: an internet outage takes the software away from every PC, so a second internet line from a different provider becomes part of the downtime plan, not an extra. Carestream also notes that sharing the hosted connection with "internet-intensive services such as video streaming or VOIP could affect WinOMS Hosted performance", which matters when the backup line carries the phones too.

Backups differ as well. The vendor does not support live backups of the WinOMS database, so ask IT how the backup schedule works around that. The trade-offs between the three deployments are in our comparison of WinOMS server vs Hosted vs Sensei Cloud.

How do you finish a sedation case on paper? The fallback kit

You finish it on the paper form your anesthesia team already uses, with the readings taken from the monitor display, and you enter or scan that paper into the record once the system is back. IT's job is to make sure the forms exist, are the current version, are in every room and that someone knows how to get them back into the software.

A paper anesthesia record only helps if it is in the room when the screen freezes. That is what a downtime box is for: one in each surgical suite and one at the front desk, checked every surgery morning.

The contact card matters as much as the forms. CISA advises keeping "a hard copy of the plan and an offline version", and SAFER suggests "a job aid, such as a small, self-contained reference card or checklist." If the phone numbers you need live only in the practice software or in email, you lose them at the moment you need them.

Getting paper back into the record

Downtime ends twice: once when the system comes back, and again when the paper is in the record. Plan the second ending as carefully as the first.

1. 1. Number every form. Each paper form gets the number from the downtime log, so nothing is missed or entered twice.
2. 1. Name one person to enter. One person (or one per suite) enters or scans the paper into the record, so there is a single owner for each case.
3. 1. Reconcile monitor data. If your monitor stores readings, check with the manufacturer or your biomedical service how to retrieve them; do not change monitor settings to do it.
4. 1. Mark the record. Note in the record that the case was documented on paper during downtime and when it was entered.
5. 1. Close the log. The downtime lead checks every numbered form against the log before the envelope is filed.

Hospitals show why this step deserves an owner. Larsen and colleagues (Applied Clinical Informatics, 2019) studied electronic health record downtime at two mid-Atlantic hospitals. At one of them, a 300-bed suburban hospital, a 2016 ransomware attack caused "approximately 48 hours during which all systems were completely shut down, followed by 48 hours of partial downtime" while systems were restored (Methods, Location). The authors also note that hospital contingency plans often leave total downtime out, because "typically it is assumed that at least a read-only snapshot of the EHR would remain accessible", and that a deliberate cyberattack can remove even that copy (Background and Significance). In the same study, laboratory results were delayed by an average of 62% compared with normal operation, and the archival data were incomplete "due to inconsistencies in the downtime paper records" (Results). These are figures from one hospital study, not dental figures. A small office has fewer forms, but the same gap opens if nobody owns the back-entry, and a plan that assumes a read-only copy will survive should be tested without one.

What should you check every surgery morning? A 30-minute pre-flight

Thirty minutes before the first patient, named owners check six things: the software and phones, last night's backup, every suite's workstation and monitor link, the network closet, the downtime boxes and the contact card. Each line below has one owner, so a gap is found at 7:10, not at 8:40 with a patient sedated.

Carestream's requirements say "wireless networking is not recommended for Anesthesia", which is why the 07:10 line checks the cable, not the Wi-Fi bars. If the referral line runs on an old copper phone line, talk to your phone provider or see our phone systems for dental offices page; a line that stops ringing is a downtime of its own.

The weekly, monthly and quarterly add-ons

Print the scenario table and this pre-flight together, fill in your office's names and numbers, and put a copy in each downtime box.

How much downtime can you afford? Set your restore point and restore time

Two numbers decide how your backups and failover should be set up, and only you can supply them. The restore point is how much recent work you could re-create if the server died now; the restore time is how long you can run on paper before cases have to be cancelled.

Restore point (RPO) is how much recent data you can afford to lose. Restore time (RTO) is how long you can run without the system. In an oral surgery office, set the restore point by how much charting and anesthesia documentation can be re-entered from paper, and the restore time by how many sedation cases you would have to cancel.

Work through these with your own figures:

1. 1. Restore point. "If the server failed at 11:00 a.m., how many hours of charts, anesthesia records and images could we re-create from paper and from the monitor?" If the honest answer is "none after this morning's first case", a nightly backup alone is not enough. The vendor requires the database engine to be stopped for a proper copy, so ask IT how protection during the day works within that limit.
2. 1. Restore time. "How many sedation cases would we cancel for each hour the system is down, and what does each one cost us in rescheduling, staff time and patient goodwill?" Use your own case count and your own figures; the point is to know the number before the outage, not to guess it during one.
3. 1. What the restore actually needs. List what has to come back for the office to run: the database server, the imaging store, the monitor-station PCs, the firewall and the internet line. Each one needs a restore path and a person.

A UPS worksheet, without guessing minutes

Battery runtime depends on your UPS model and what you plug into it, so this page gives the method, not a number:

1. 1. List each device on the UPS: the server, firewall, switch, internet modem, the monitor-station PC, the phone system's power-over-Ethernet switch.
2. 1. Add up their power draw from the device labels, or ask your IT provider for the figures.
3. 1. Read the UPS maker's runtime chart for that total.
4. 1. Compare the result with the time it takes to finish recording the current case plus a clean server shutdown. If the battery is shorter, move devices to a second UPS or change the unit.

Carestream recommends "a UPS with shutdown software installed with the server machine." The software is the part that turns a power cut into a clean shutdown instead of a damaged database. Our dental backup and disaster recovery page describes the backup side of this work.

What do recent oral surgery data breaches teach about downtime?

They show that the systems that go down are not the only systems at risk. In three cases reported publicly in 2025 and 2026, patient data was exposed through imaging files, through files stolen without anything being encrypted, and through a vendor's email. A downtime plan that only covers restoring the server misses all three.

The question many offices ask is "what is the biggest data breach threat for us right now?" For an oral surgery practice, the answer starts with what the office holds and where it lives.

The oral surgery data profile

How referrals and their images arrive is its own process; see our guide to setting up oral surgery referral intake.

Three cases, facts then lesson

The facts below are as reported by The HIPAA Journal, with the dates shown. They are summarised to draw IT lessons; nothing here suggests who provided IT to these practices or who was at fault.

1. West Texas Oral Facial Surgery, Lubbock, Texas (reported 11 August 2025). According to The HIPAA Journal's report, the practice experienced network disruption on May 29, 2025, and notified 11,151 patients. Exposed data included names, imaging files (in some cases with birth dates) and the reason for seeking treatment. "The electronic medical record system was not accessed." The Inc Ransom group claimed the attack. This is the oral surgery ransomware case to remember: the medical record system was not accessed, but imaging files were exposed. Lesson: back up, segment and monitor the imaging store as carefully as the practice-software server. How to separate the surgical and imaging networks is covered in our guide to connecting sedation monitors and segmenting the surgical network.

2. Arkansas Oral & Maxillofacial Surgeons, Hot Springs, Arkansas (reported 12 August 2026). According to The HIPAA Journal's report, the incident was first identified on April 7, 2026, and on June 2, 2026, it was confirmed that an unauthorized third party had accessed the network and taken files containing patient information. It was reported to HHS as affecting 64,831 individuals. The group that claimed it, PEAR, "does not encrypt files"; it steals data and threatens to publish it. Lesson: a perfect backup does not undo data theft. This oral surgery data breach is the reason backup is paired with MFA, endpoint protection and monitoring for unusual outbound traffic. Our dental cybersecurity services page covers that side of oral surgery cybersecurity.

3. Stafford Oral Surgery, Virginia (reported 10 June 2026). According to The HIPAA Journal's report, a third-party vendor's email accounts and files were accessed between October 15 and October 23, 2025, after a phishing attack, and Stafford Oral Surgery was listed with 7,019 affected individuals. The report says patient medical and dental records were not accessed. Lesson: the downtime and incident plan includes the vendors who hold your data. Keep their incident contacts on the printed card, and read up on email security for dental offices.

Notification duties after a breach are a legal question; ask your counsel. For the status of proposed changes to the Security Rule, see our summary of the HIPAA Security Rule changes.

What should you ask your IT provider and your software host?

Ask the twelve questions below and write down the answers. A good provider answers most of them from records it already keeps; a vague answer to a backup or restore question is a finding in itself. Questions 4 and 5 go to your software host if your system is hosted or in the cloud.

What your IT agreement should say about response and escalation is a separate question; our guide to what your IT support agreement should define covers it.

Opening a second office or a new oral surgery practice? Build the plan in

The cheapest time to build a downtime plan is before the first patient, while the rack, the circuits and the suites are still on paper. Adding a second internet line or a UPS after opening means downtime of its own. Three items belong in the build plan, not on a list for later:

• • A failover internet circuit ordered with the main line, from a different provider, so both are live on opening day.
• • A UPS with shutdown software in the rack plan, plus battery backup for each monitor-station PC and the switch it depends on.
• • A downtime box in each suite at opening, with the contact card printed before the first sedation case.

As an example of the build stage, Legend's March 2026 first look at the technology buildout for Ecker Oral Surgery in Miami covered the network rack, computers, AV systems and security surveillance. The opening sequence for any dental office is in our guide to opening a dental practice: the IT plan, and the oral surgery additions are in the oral surgery practice technology guide.