On a Tuesday afternoon, the district care network had to cover an unexpected absence. The duty phone, which receives calls from families and professionals, was with a volunteer who had suddenly fallen ill. An apparently obvious answer appeared in the group chat: hand the mobile to the first person to reach the centre. The shift coordinator saw an urgent need; the privacy lead remembered that the device displayed sensitive conversations; a care worker asked that a family waiting for a reply not be delayed. No one was arguing for insecurity or bureaucracy. Each person was protecting a different part of the same situation.
The discussion moved forward when they stopped asking who was right and asked something else: what are we protecting, and what specific inconvenience are we willing to accept in order to protect it? That question does not remove risk. It makes risk discussable before a quick measure turns a misunderstanding into routine. A risk pact is useful exactly here. It does not prescribe one universal answer; it lets people agree on a boundary, an owner, and a way to review it.
Urgency does not make access harmless#
The phone was not simply an object that rang. It held contact numbers, follow-up notes, messages from people seeking help, and the ability to answer in the network's name. Handing it over with no conditions solved the first visible problem — someone could answer — but introduced others: messages visible to bystanders, changes with no record, conversations that an occasional substitute did not need to know, or a device leaving the chain of custody.
The phone was not simply an object that rang. It held contact numbers, follow-up notes, messages from people seeking help, and the ability to answer in the network's name.
It was also true that freezing the handover until a formal meeting would have failed the people calling. The point was not to choose between total openness and paralysis. It was to recognise that both extremes transferred a cost to someone. Openness could damage the confidentiality of people asking for help; inaction could leave a family without a response. The pact turns that collision into an explicit decision rather than hiding it behind phrases such as “only for today” or “this is how we have always done it.”
The network wrote down three things it wanted to preserve: continuity of care, privacy of conversations, and the ability to explain later what had happened. It then listed acceptable inconveniences: call a named contact before handing over the phone, limit the substitute's coverage to a defined time window, and ask the coordinator to handle matters requiring prior history from her own channel. This was not a heroic document. It was a decision specific enough that no one had to improvise alone.
Opposing roles reveal the whole risk#
One difficulty with any control is that the person proposing it tends to see mainly the harm they want to prevent. The network therefore held a short session with deliberately opposed roles. The coordinator argued for continuity: “If it rings and nobody answers, we are failing too.” The privacy lead raised scope: “The cover person does not need to read weeks of messages to return a call.” The care worker represented a family's experience: “I do not want to repeat my story to three people, but I do not want just anyone to see it either.” A logistics volunteer asked what everyone had missed: “Who collects the phone after the shift, and how do we know it was handed back?”
The purpose was not to vote for a winning position. It was to discover assumptions. Continuity did not require unlimited access; confidentiality did not require leaving the mobile switched off; making things easy for the cover person did not justify leaving them without guidance. When roles are heard separately, a more practical answer emerges than any general slogan: a substitute can take new calls and log a message, but cannot search old conversations unless the coordinator authorises and accompanies the lookup.
The roles change the tone as well. Instead of accusing somebody of being careless or inflexible, each participant defends a legitimate need that the group must combine. That distinction matters. Controls that feel like punishment are bypassed; controls born from a shared explanation can be adjusted when reality tests them.
Designing a handover with visible boundaries#
The final agreement had five parts. First, a short handover record: time, the name of the recipient, coverage window, and the person retaining escalation responsibility. Second, a lock screen and messaging account that did not reveal full history in notifications. Third, a two-page guide with basic responses, emergency numbers, and one clear rule: do not copy messages into personal chats. Fourth, an immediate consultation route for cases requiring earlier information. Fifth, a closing time to return the device and confirm that no notes had been left outside the record.
None of these elements guarantees that an error will never occur. Together, they reduce the chance that an ordinary mistake becomes a wider exposure. More importantly, they make covering the duty phone feasible for someone who does not know every detail of the service. A good control does not assume that a new person will remember an extensive policy under pressure. It prepares a safe path that still allows action.
The network also decided which inconvenience it would not accept. It would not ask the substitute to use the ill volunteer's personal password, send screenshots of conversations as a briefing, or leave the phone in a drawer accessible to anyone at the centre. Naming these boundaries mattered because it prevented informal answers that might have seemed reasonable in the middle of fatigue. What is not agreed can reappear as an exception the next day.
Reviewing the decision is also care#
At the end of the shift, the group neither looked for blame nor declared victory. It reviewed three facts: how many calls arrived, which questions the substitute could not resolve, and whether any step caused an unjustified delay. They found that the guide needed a clearer distinction between an orientation call and an alert that should go immediately to a professional. They also found that the handover record was buried too deeply in a folder. They changed both things before the next shift.
That review is part of the pact, not an audit imposed from outside. Accepting an inconvenience is reasonable only if the group checks that it remains proportionate to the harm it prevents. If every handover takes twenty minutes and does not improve protection, the procedure needs correction. If a shortcut keeps returning because the guide fails in real situations, the shortcut is a design signal, not proof of bad intent.
It is useful to record the reason for the choice as well. A plain note might say: “History access was limited to protect privacy; new calls remained covered by a substitute; review on Friday.” Someone joining the team then does not inherit a mysterious rule. They inherit a decision with purpose, scope, and a review date. That memory matters especially in networks built on shifts, substitutions, and trusted relationships.
The next agreement can be made before a crisis#
There is no need to wait for illness or an urgent call to prepare this kind of pact. At its next meeting, an organisation can choose one resource that passes from hand to hand — a phone, shared account, or digital key — and ask four questions: which service cannot stop, which information does not need to circulate, which inconvenience will we accept, and who decides and reviews an exception? The answers do not have to be technical. They must be usable when time is scarce.
The duty-phone case left a modest lesson. Protection does not mean banning handovers; it means refusing to turn urgency into unlimited permission. A risk pact gives that boundary a shape. It lets the network respond to callers, care for people who shared information, and give the substitute a task they can perform with confidence. When the next unexpected event comes, they will not have to choose between responding and protecting. They will have an agreement for doing both responsibly.
