7 Signs Your Hospital's "Managed" PACS Support Isn't Actually Managed
Managed PACS support red flags: slow SLAs, fake 24/7 cover, vendor lock-in and unowned interfaces — with what each one should look like instead.
The phrase managed PACS does a lot of unearned work in imaging IT contracts. Almost every support agreement now carries the word, and a large share of them describe something considerably closer to break-fix maintenance with a nicer cover page. The distinction only becomes visible on a bad night — which is a poor time to discover it. Here are seven signs the arrangement is not what the contract calls it, each tied to the requirement it is failing.
per minute — worst-case unplanned healthcare downtime cost found by the Ponemon Institute
downtime cost at one regional radiology network, per downtime analyses reported by AuntMinnie
typical cost of a single 6–8 hour MRI downtime event to a radiology department
estimated industry-wide cost of healthcare IT downtime
1. Your escalation path ends at a queue, not a person
You log a severity-one incident and receive a ticket number. Two hours later you have an acknowledgement and no owner. The requirement being missed is a named escalation matrix: for each severity tier, a specific engineer, a specific response window and a specific fallback if that window is breached. If nobody's name appears anywhere in your support documentation, the contract is a queueing arrangement.
2. "24/7 PACS support" means someone answers the phone
This deserves its own section, and gets one below. The short version: answering a phone is not support. If no engineer can act until your team notices a problem and reports it, you have an out-of-hours contact number, not overnight coverage.
3. Nothing is monitored — everything is reported
In a genuinely managed environment, the provider tells you about the archive volume approaching capacity, the interface that stopped acknowledging at 11 PM, or the modality that has not sent a study in six hours. In a maintenance environment, you find these things yourself, usually via a technologist who cannot complete a study. Continuous monitoring with alerting on storage, interfaces, routing and service health is the dividing line between the two models.
4. The same interface breaks every quarter
Recurrence is the clearest tell of all. A fixed incident that returns was never root-caused; it was restarted. Managed support includes problem management as distinct from incident management — a documented root cause, a permanent remediation and a change record. Because HL7 and DICOM integration points are where most imaging IT faults originate, this is also where recurrence hurts most. That whole layer, from DICOM gateway routing and connectivity outward, should have a named owner.
5. Your support is tied to one PACS brand
A vendor-tied support model has a structural incentive that is not aligned with yours: problems tend to resolve toward the product roadmap. It also leaves the multi-vendor reality of a real estate — a VNA here, a legacy archive there, a departmental system somewhere else — formally out of scope. Vendor-neutral support covers what you actually run and coordinates with the OEM on your behalf. The RAD365 PACS support framework sets out how those coverage tiers are defined.
6. Nobody can tell you what the environment looks like
Ask your provider for a current interface inventory, an AE title map and an archive growth projection. If producing it takes three weeks, the environment knowledge is not documented anywhere — it lives in one person's memory, and that person can resign. This is the single largest hidden risk in imaging IT, and it is the reason interim PACS support exists as a distinct service: somebody has to hold the environment while the knowledge is rebuilt.
7. The SLA has no consequence and no percentile
An SLA that reports tickets closed, with no measurement against target and no remedy for a breach, is a description rather than a commitment. What a real SLA reports: response and resolution times by severity tier, measured against contracted targets, with percentile figures rather than averages, split between business hours and out-of-hours, on a fixed reporting cadence.
Stat callout
The Ponemon Institute has measured worst-case unplanned healthcare downtime at $17,244 per minute. At that rate a single hour costs more than most annual support contracts — which is why response-time SLAs are a financial control, not an administrative detail.
24/7 PACS support: what the phrase should actually mean
24/7 PACS support is not a high-volume search term and we are not going to pretend otherwise — it is an underused exact phrase that describes something buyers genuinely need to specify, so it is worth defining precisely rather than leaving to interpretation. Four components have to be present:
- Continuous monitoring. Automated checks on archive, interfaces, routing and service health, running overnight and at weekends, with alerting that reaches a human.
- A staffed rota. Engineers on shift or on a genuine call rota at 3 AM — not a message service that pages someone who will look in the morning.
- Out-of-hours SLAs stated separately. If the contract quotes one response time without distinguishing 2 PM from 2 AM, assume the 2 AM figure is worse.
- Holiday and weekend parity. Imaging volume does not observe public holidays, and neither should the coverage standard.
Everything short of that is business-hours support with a phone number attached. The operational layer that sits behind real round-the-clock cover is described further in radiology IT support.
Business hours versus genuine 24/7/365
| Capability | Business-hours support | 24/7/365 managed support |
|---|---|---|
| Overnight monitoring | None | Continuous, with alerting |
| 3 AM response | Answering service | Engineer, contracted window |
| Interface ownership | On request | Proactive, with problem management |
| Storage oversight | Reactive at capacity | Forecast and planned |
| OEM coordination | Facility does it | Provider does it on your behalf |
| Reporting | Tickets closed | Percentiles against SLA targets |
A five-step review you can run this month
- Pull twelve months of tickets and count recurrences by system and interface. Recurrence rate is your problem-management score.
- Separate out-of-hours incidents and measure response times against the daytime figure. The gap is your real coverage picture.
- Request the environment inventory — interfaces, AE titles, archive growth. Time how long it takes to arrive.
- Read the SLA for consequences. If a breach has no remedy, it is a target, not a service level.
- Cost one bad night at your own volumes using the published downtime figures, then compare it with your annual contract value.
Where that review exposes a platform problem rather than a support problem, the conversation shifts to PACS migration services; where it exposes an administrative-capacity problem, to radiology administrative support. Most of the time it exposes neither, and the fix is simply a support model that matches the operation.
About the author
Trisha Seal writes on imaging IT and PACS operations for RAD365. RAD365 provides vendor-neutral managed and outsourced PACS support — continuous monitoring, DICOM and HL7 interface management, archive and storage oversight, OEM coordination and severity-tiered SLAs — for hospitals, imaging centres and multi-site groups.
Pressure-test your PACS support contract
Talk to a RAD365 PACS engineer about coverage tiers, out-of-hours SLAs and what your current contract actually obliges anyone to do.
Review our managed PACS coverage →Managed PACS support: frequently asked questions
Recognizing the Warning Signs
What are the most common signs a hospital has outgrown its current PACS support?
Tickets that get acknowledged but not owned. Interface breaks that recur on the same interface every quarter. Storage decisions being made reactively when a volume fills. A single internal administrator who cannot take annual leave without risk. And escalation paths that end at a queue rather than a named engineer. Individually these look like irritations; collectively they describe a support model that is reacting to an imaging operation rather than running alongside it.
How can you tell if "24/7 PACS support" is real or just marketing language?
Ask three questions. Who answers at 3 AM on a Sunday — an engineer or an answering service? What is the contracted response time at that hour, and is it the same as the daytime figure? And is anything actually being monitored overnight, or does the clock only start when your technologist notices and calls? Real round-the-clock cover means continuous monitoring plus a staffed rota. Marketing round-the-clock cover means a phone number that is answered outside business hours.
What does it mean when a PACS vendor calls their service "managed" but response times are slow?
It usually means the contract covers the software product rather than your environment. The vendor fixes defects in their application; nobody is watching your storage headroom, your HL7 interfaces, your DICOM routing or your integration points. That gap between "the software works" and "the imaging operation works" is where slow response times actually live, and no amount of relabelling closes it.
How often should a hospital re-evaluate its PACS support contract?
Annually as a discipline, and immediately on any of four triggers: a change of PACS platform or version, a merger or new site, the departure of the internal administrator who held the environment knowledge, or an outage that exposed a coverage gap. Contracts written for a single-site environment quietly stop fitting once the estate grows, and nobody re-reads them until something breaks.
What Managed PACS Support Actually Includes
What's the difference between managed PACS support and basic PACS maintenance?
Maintenance is reactive and product-scoped: patches, version updates, defect fixes. Managed support is proactive and environment-scoped: continuous monitoring, storage and archive oversight, interface and integration management, change control, vendor coordination on your behalf, and documented SLAs with consequences. Maintenance keeps the software current. Managed support keeps the imaging operation running.
Does managed PACS support cover DICOM connectivity and interface troubleshooting?
It should, and if it does not the contract has a hole in exactly the place most incidents occur. Modality-to-archive routing, AE title and configuration management, HL7 order and result interfaces, worklist behaviour and downstream integrations are where the majority of day-to-day imaging IT faults originate. RAD365 covers this layer explicitly, including DICOM gateway configuration and connectivity.
What SLAs should a managed PACS support company guarantee?
Severity-tiered response times with distinct targets for a total archive outage, a broken interface and a single-workstation issue; an out-of-hours target that is stated separately rather than assumed identical to daytime; a named escalation path with a person at the end of it; and reporting against those targets on a schedule. An SLA that reports only on tickets closed, with no percentile against target, is not measurable.
Is managed PACS support vendor-neutral, or tied to one PACS brand?
It can be either, and the difference matters enormously. A vendor-tied model has a structural incentive to route problems toward its own product roadmap. A vendor-neutral provider supports whatever you are running, coordinates with the OEM on your behalf, and can advise on migration without a conflict. RAD365 is vendor-neutral — see the PACS support framework for how coverage tiers are structured.
What's included in 24/7/365 managed PACS support versus business-hours-only support?
Business-hours support gives you engineers between roughly 8 and 6 on weekdays; everything outside that waits or goes to an answering service. Genuine 24/7/365 support adds continuous automated monitoring with alerting, a staffed out-of-hours rota with contracted response times, holiday and weekend cover at the same standard, and overnight incident handling that starts before your team notices. Since imaging runs around the clock, the second model is the one that matches the operation.
Cost and ROI
How much does managed PACS support typically cost compared to an in-house administrator?
The honest comparison is not salary versus fee. A single in-house administrator is one person with one skill set who sleeps, takes leave and eventually resigns, taking undocumented environment knowledge with them. A managed contract buys a team, out-of-hours cover, monitoring infrastructure and continuity. Set the true loaded cost of the role — plus the cost of the coverage it does not provide — against the contract before concluding which is cheaper.
What does unplanned PACS downtime actually cost a hospital per hour?
The published figures are sobering. The Ponemon Institute has found unplanned healthcare data-centre and system downtime costing as much as $17,244 per minute in worst-case incidents. Downtime-cost analyses reported in AuntMinnie's coverage of the true cost of downtime in radiology cite figures such as $40,000 per hour at one regional radiology network, and MRI downtime events averaging six to eight hours costing a department $50,000 to $75,000 per incident. Even the conservative end of that range dwarfs a support contract.
Is outsourcing PACS support cheaper than hiring a full-time PACS administrator?
Frequently, and more reliably so once out-of-hours coverage is priced in — because matching a managed rota with employees means hiring several people, not one. The stronger argument is risk rather than price: broader healthcare IT downtime is estimated at roughly $1.9 million per day industry-wide, and a single avoided multi-hour archive outage can pay for a year of coverage. Many organisations run a hybrid, keeping an internal owner and contracting the depth and the night shift.
Choosing and Switching Providers
What questions should a hospital ask before switching managed PACS support providers?
Who are the named engineers assigned to our account? What are the severity-tiered response times, including overnight, and how are they reported? What is monitored continuously, and what only generates a ticket when we call? How is OEM coordination handled? What does onboarding cover in terms of environment documentation? And what happens to that documentation if we leave?
How disruptive is it to migrate from one managed PACS support provider to another?
Considerably less disruptive than a platform migration, because the systems do not move — only the support layer does. The risk sits in knowledge transfer. A structured transition means a documented environment discovery, mapped interfaces and integrations, an agreed escalation matrix and a parallel-running period before handover. Where the outgoing provider is uncooperative, discovery is the work that protects you; interim PACS support exists to cover exactly that window.
What certifications should a managed PACS support company hold for healthcare IT compliance?
Look for demonstrable HIPAA compliance including a signed BAA, documented information-security controls and audit logging, evidence of secure remote-access practices, background-checked engineers, and change-control and incident-management processes you can inspect. Ask to see the actual artefacts. A certification claim without an auditable process behind it protects nobody.
Can managed PACS support work across multiple imaging modalities and EHR systems?
Yes — that is a core requirement rather than an add-on. A working model supports CT, MR, US, CR/DR, mammography, nuclear medicine and any other modality on the estate, with HL7 and API-level integration into whichever EHR is in place, plus VNA and cloud archive layers. Multi-site groups typically run mixed estates, which is precisely why vendor neutrality and integration breadth matter more than familiarity with a single brand.