An ITOps Audit Checklist for Hospital Administrators

Health Technology

This is a working checklist, not a lecture. It is meant for the administrators and operations coordinators who keep a hospital-side office running: scheduling systems, records portals, billing screens, internal messaging. Clinical decisions and patient care are out of scope; the list covers only the supporting technology and the routines around it. A majority of “yes” answers means your ITOps is in decent shape. Every “no” is a small job to schedule.

The questions are drawn from the way the Ops Atlas describes ITOps, the function that keeps IT services running, and they are grouped so you can audit one area per week.

Area A: Requests and incidents

  • Is there one named place, a mailbox or a form, where staff report technical trouble?
  • Does every report get a reference number and an owner within the same working day?
  • Can you tell, without asking around, how many reports were open at the end of last week?
  • Are routine requests, such as a new staff login or a replacement keyboard, handled by a repeatable checklist instead of an improvised email thread?
  • Do incidents and routine requests live in separate lists, so urgent breakage is not buried under supply orders?

The Atlas treats the service desk, incident handling and request fulfilment as three separate practices for a reason: they carry different urgency. Mixing them is the quickest route to missed fixes.

Area B: Repeat problems

  • Does anyone review the log monthly for entries that appear more than once?
  • Does each repeated issue have a named person looking for the cause?
  • Are workarounds written down so a colleague on another shift can apply them?

A preparation habit makes a good comparison. The suggestions in this article on getting ready for a checkup so the visit goes smoothly are about gathering what you need in advance. A repeat-problem review does the same for IT: it collects what you know before the next failure arrives.

Area C: Changes

  • Is there a written definition of what counts as a change to a live system?
  • Do you distinguish routine changes that are pre-agreed from those that need an assessment, and from urgent fixes?
  • Does the assessment use a written risk score, even a three-line one?
  • Is there a shared calendar showing planned work and quiet periods, such as month-end billing?
  • Do you keep a record of every change, including who approved it and when?
  • Is each failed or urgent change reviewed afterwards?

The Atlas stresses two points that surprise newcomers. Pre-authorised routine changes exist precisely so that not everything needs a committee. And the DORA research programme it cites found no evidence that outside approval lowers failure rates, recommending peer review instead. If you want your own team to practise that peer habit, ask a colleague to read every non-routine change plan before it goes ahead.

Area D: Monitoring and alerts

  • Do you monitor the handful of things that would stop the front office: logins, scheduling, billing and messaging?
  • Does a human actually receive the alerts, and do they know what to do?
  • Do related alerts get grouped, so one failure creates one ticket?
  • Have you tested the alert path in the last month?
  • Is the alerting system itself treated as something that must stay up?

For the grouping question, the Atlas describes an exercise where thirteen alert rows condensed into two incidents inside a time window. The principle to take from it is that volume is not the same as importance. And its warning about the alerting layer is blunt: when it fails silently, outages stay hidden.

Area E: The inventory

  • Is there a one-page list of the systems whose failure would stop work?
  • Does each system have an owner, a supplier contact and a renewal date?
  • Is the list reviewed at least quarterly, with stale entries flagged?
  • Do shadow spreadsheets exist alongside the official list, and if so, why?

The Atlas warns that an inventory crammed with everything is less useful than one holding what decisions need. In its sample of ten records, nine were complete, but two were stale, one had no relationships and one repeated a serial number. Use that as a yardstick: if you audit ten of your entries and find four or more issues, you are in familiar territory.

Area F: Promises and clocks

  • Are repair targets written for each priority level?
  • Do the rules state which hours count and which days pause the clock?
  • For targets that depend on suppliers, do you know the supplier’s own commitment?
  • Is there a buffer between the sum of the parts and the promise you make to staff?

Here the Atlas illustration is helpful: an 8-hour commitment is carved up as 30 minutes of triage, 2 hours for a network check, 4 hours owed by a carrier and 1.5 hours left unallocated.

Area G: Knowing the territory

  • Has someone read a neutral reference on the discipline? A compact place to start is the Atlas’s ITOps entry.
  • Does the team agree on which problems belong to IT operations and which belong elsewhere, such as external customer support?
  • If a legal or contractual obligation touches your systems, has a qualified adviser confirmed what applies? This checklist cannot do that for you.

Scoring yourself

Count your “yes” answers. Twenty-six or more out of the thirty questions suggests a mature routine. Between eighteen and twenty-five, you have solid foundations with some gaps. Below eighteen, choose two areas and work only on those for a month. The scoring bands here are my own suggestion, not an official scale, and exist only to help you prioritise.

Turning the audit into action

  1. Print the checklist and mark each question yes, no or unsure.
  2. Pick the three “no” answers that would cause the most disruption.
  3. Give each one a named owner and a deadline.
  4. Repeat the audit after a quarter and compare the scores.

People new to the field sometimes use an audit like this as the backbone of an interview story. If that is your goal, the practice plan in a roadmap for preparing for technical interview rounds pairs well with it.

Keep the checklist short enough to use, honest enough to be uncomfortable and regular enough to matter. That is all ITOps asks of an administrative team.