NEMT dispatch software down, phones out, or no cell service: running the day on paper

What a real outage looks like

Not every outage comes with a storm warning. On February 22, 2024, an AT&T employee added a misconfigured network element during routine overnight maintenance. Three minutes later, at 2:45 a.m. Central time, AT&T’s wireless network disconnected every device on it. The FCC’s July 2024 report on the outage found:

• More than 125 million devices lost voice and 5G data service, in all 50 states, Washington, D.C., Puerto Rico, and the U.S. Virgin Islands.
• More than 92 million calls were blocked, including more than 25,000 calls to 911.
• Customers of other carriers who were roaming on AT&T’s network, or using a carrier that runs on it, lost service too.
• Rolling back the change took close to two hours. Full service took at least 12 hours, because the systems that reconnect phones were swamped by devices trying to get back on.

The FCC called it a “sunny day” outage, meaning it had nothing to do with weather. For a NEMT fleet on that carrier, it began before the first pickups and ran into the afternoon.

Outages also hit the other side of the trip. On May 22, 2024, Call the Car, the transportation vendor for L.A. Care Health Plan, hit a disruption during a routine backup that pointed to possible malware or ransomware. It moved to a backup system with limited features, members had trouble with its app, and calls to book rides took longer. Your plan has to cover your own systems, your phones, and your brokers’ systems.

The written plan HIPAA expects

Any NEMT company HIPAA covers, directly or as a business associate, needs a contingency plan under the Security Rule. The standard at 45 CFR 164.308(a)(7) covers any emergency or event that damages systems holding electronic patient information, and lists fire, vandalism, system failure, and natural disaster as examples. It has five parts:

1. Data backup plan (required). Retrievable exact copies of the electronic patient information.
2. Disaster recovery plan (required). Procedures to restore any lost data.
3. Emergency mode operation plan (required). Procedures to keep critical work going, and keep the information protected, while operating in emergency mode.
4. Testing and revision procedures (addressable). Periodic tests and updates of the plan.
5. Applications and data criticality analysis (addressable). Which systems and data matter most.

Section 164.316 requires these policies in written form, which may be electronic, kept six years past the later of the day they were written or the day they were last in effect. The HHS paper on administrative safeguards suggests what a small company can ask itself: whether the emergency plan includes manual procedures for protecting patient information, whether it lists the phone numbers and names of everyone to contact, and whether a copy is kept in more than one place. It also says a scenario walk-through is an acceptable way to test. The HIPAA guide for NEMT explains which NEMT companies HIPAA covers directly and which ones it reaches as business associates.

A January 2025 proposal to update the Security Rule would add a written duty to get critical systems and data back within 72 hours of losing them. As of October 2026 it remains a proposed rule, with no final rule published.

Broker contracts ask for the same thing in plainer words. MTM’s standard provider agreement (the January 2023 copy posted by Pennsylvania) requires an office that can respond right away during business hours, working computers, software, and internet, a contingency or backup plan for every accepted trip, and two-way voice contact with every vehicle. In Virginia, MTM Health’s handbook has each driver carry a working phone or similar device all through service hours, so the driver can always reach dispatch and MTM. Modivcare’s 2025 compliance training for its transportation providers lists a contingency plan for system emergencies and natural disasters among its HIPAA best practices. Under CareOregon’s provider manual, each brokerage and every provider it contracts with must have business continuity and disaster recovery policies in writing that address staff training, notice methods, and member education at a minimum, with copies supplied when asked.

Match the fallback to what failed

Each kind of failure leaves different tools working. Decide the first move for each one before it happens.

Two lessons from the 2024 outage shape the phone plan. AT&T’s own advice that morning was to use Wi-Fi calling, though some users got errors when they tried to turn it on. And roaming customers of other carriers lost service as well, so “a second carrier” has to mean a carrier that runs its own network. Put at least one phone in the office and one in each zone on that second carrier.

Emergency alerts kept working. The FCC found that AT&T’s 4G network still broadcast Wireless Emergency Alerts through the outage, so a driver’s phone could still warn of a tornado even when it could not place a call. Calls to 911 were another matter: phones in SOS mode could reach 911 through another carrier’s tower, but not through AT&T’s. A driver with a rider in distress needs a second way to call for help, such as the clinic’s landline.

The paper kit to print every evening

The paper plan only works if the paper exists before the outage. Print it at the end of every day, for the next day:

• The manifest. In Virginia, MTM Health’s manifests carry everything a driver needs without a screen: a unique trip ID; the rider’s name and phone number; where and when the pickup and appointment are; the doctor or other provider, with a phone number; the approved mode and how much help the rider gets; whether an escort comes along or your company sends an attendant; the rider’s weight including any mobility device or equipment; and special instructions. Print it in trip order for each driver. A daily manifest template shows a layout.
• The contact sheet. Each driver’s phone, each broker’s provider and trip lines, the dialysis centers and facilities you serve, your backup vendor, and the steps to forward the office line.
• Blank trip logs. A stack in each van, with a line for every field your brokers need to pay. The trip log template is a starting point.

Paper manifests hold patient information, so handle them that way. Modivcare’s 2025 provider compliance training asks providers to keep trip records where others cannot see them, not to leave patient information in unlocked vehicles, and to dispose of paperwork properly. HHS lists shredding, burning, pulping, or pulverizing as proper disposal for paper records. Collect the sheets at the end of the shift and shred them once the trips are entered.

Writing trip times by hand so the trips still get paid

The handwritten log has to carry every field the broker checks. Under MTM Health’s Virginia handbook, a trip goes unpaid if its electronic log lacks any of these:

• the trip ID
• both pickup times, scheduled and actual
• when the van left with the rider and when it arrived
• the member’s signature

The same handbook requires vehicle location tracking for the whole ride. A paper log cannot replace that by itself, so call the broker’s provider relations line on the day of the outage. Give the hours affected and the trips involved, and ask how they want those trips documented. Get the answer in an email.

On the road, drivers write each time from a watch or the dashboard clock as it happens, along with odometer readings at pickup and drop-off, and get the rider’s signature on the paper log. Never fill in times afterward from memory. Federal Medicaid rules require providers to keep the records needed to show how much service they provided (42 CFR 431.107), and an audit will compare those records with the claim.

Notify brokers first, then facilities and riders

Notice protects the contract. Under MTM’s standard agreement, a significant delay that will make a member late for an appointment has to be reported to MTM right away, along with alternate plans to finish the trip if the appointment can still be kept. Work the list in this order:

1. Brokers. Tell each one what failed, when it started, and which trips are at risk, and ask for its outage instructions.
2. Facilities. Call dialysis centers and clinics with their patients’ pickup status.
3. Riders. Call the riders whose times changed, and anyone who does not use text.

When the outage is on the broker’s side, follow its notices. During the 2024 vendor outage, L.A. Care asked offices to book within the plan’s advance-scheduling windows and not too far ahead, to keep phone lines open for urgent calls.

Catching up once systems return

Enter the day the same day. Memory fades fast, and brokers set filing limits: MTM Health’s Virginia handbook gives providers six months from the trip date to submit a clean claim, but the trip still has to match the log.

1. One person enters the paper logs. A second checks each entry against the paper.
2. Mark the outage window in your records so an auditor can see why those trips were logged by hand.
3. Send the broker the documentation it asked for.
4. Shred the paper once entries are checked, unless your broker or records policy says to keep the originals.
5. Within a week, walk through what happened with the staff involved and update the written plan. That review is the revision half of the testing and revision step HHS describes.

The emergency preparedness guide covers the weather side of keeping the office running, and the cybersecurity guide covers ransomware and backups.

When the signal drops

HealthRide’s driver app keeps working without cell service. Drivers keep recording pickups, drop-offs, and signatures in a dead zone or during a carrier outage, and the records sync once the signal returns. Each trip keeps its timestamps, signatures, and GPS-recorded miles, so the day can be checked against the broker’s records later.