Luscii - Notice history

Update:

The issue has been resolved. If the Happitech software is unable to generate the report, the measurement is no longer saved and the app asks the patient to repeat the measurement (same as on iPhones). This way, no measurements without a report end up in the Luscii patient record.

The fix is included in the latest app version: Android 5.75.0. iPhones were not affected.

Resolving the incidental failure to generate the report sits with Happitech, in their own software. We have informed them and are following up. Thank you for your patience.

We have identified the cause. The report is delivered by our measurement partner Happitech and in incidental cases it is not included. This affects a very small share of all Happitech measurements.

On iPhones this is already handled well: if the report is not delivered, the measurement is not saved and the patient can simply take a new measurement straight away. So a measurement without its report never ends up in the patient record. On Android devices, in those incidental cases, the measurement is correctly saved but without the report, which is exactly what we have seen.

We are therefore aligning Android with iOS: if the report is missing, the measurement is not saved and the patient simply takes a new measurement. That way, a saved measurement is always complete. We will let you know once that change has been rolled out.

In addition, the structural fix sits with Happitech: they need to resolve this in their own software. We have raised this with them as urgent and are actively following up.

Update:

Our investigation has shown that what initially appeared to be a single problem in fact had two separate causes. Both occur only in the Android app and only for patients enrolled in more than one programme at the same time; neither occurs on iPhone.

Importantly, no data has been lost. Every measurement was recorded - in a small number of cases it was simply filed under the wrong programme belonging to the same patient.

The first pattern involves scheduled measurements and is very rare, affecting a small number of patients. The second pattern involves separate, self-added measurements and affects a somewhat larger but still limited group of patients.

We have implemented a fix so the issue can no longer occur, and we are making additional improvements in the next app release. If this issue has occurred within your organisation, Luscii will contact you about it.

We are currently mapping the definitive scope. So far this still appears to concern a small number of isolated cases, affecting very few patients. In addition, we have identified a possible scenario that may explain the issue. As soon as we know more, we will post an update.

Update:

The support customer portal is available again. For now, you will only see tickets linked to your own email address. Tickets from colleagues are not visible. Submitting tickets through our other channels remains possible.

We are currently working on improving our support systems internally and will include visibility of organisation-wide tickets as a requirement.

This incident has been resolved. the instrument was updated to a working version again.

Update: Everything is stable and we are closing this incident. Should you notice anything unexpected, please don't hesitate to let us know. Thanks again for your patience.

For a number of care pathways, we saw that overdue alerts were not being processed automatically. This concerns a limited set of care pathways and is not the broader delay from last Thursday.

We have now found the underlying cause: during an earlier change, the correct setting was not carried over properly for a number of care pathways, which is why automatic processing was not working there. This has now been structurally fixed for the affected care pathways.

The open overdue alerts will unfortunately need to be closed manually this one time, the alerts themselves can simply be processed. We will continue to monitor this closely. We apologise for the inconvenience.

Update: The automatic processing of overdue alerts unfortunately experienced problems again today. This has now been fixed and we are monitoring the situation. We will continue to keep an eye on it over the weekend.

Unfortunately, some overdue alerts that should have been processed automatically were not, these will need to be closed manually by caregivers. We apologise for this inconvenience.

The issue has been resolved. A large volume of background tasks this morning caused a delay in processing. This was most noticeable in the forwarding of alerts to the EPD (HiX), and also affected (automatic) workflows and additional measurements. Everything has now been fully processed and the alerts in HiX have caught up, no data has been lost. We have scaled up capacity and moved the tasks in question so they no longer hold up regular processing. We are also taking steps to ensure this does not happen again. Thank you for your patience.

Update: The fix is live and the issue has been resolved - no new cases are occurring. The patients who were previously affected have also been corrected: their open alerts have now been sent through to the HiX worklist. Only alerts that had already been closed in the meantime by care providers can no longer be sent through. Thank you for your patience and our apologies for the inconvenience.

Update: Our developers have prepared a fix that resolves the issue properly. It is currently being tested. Once it has been rolled out, we will ensure that existing patients enrolled in multiple programmes also send alerts to the worklist again. We know exactly which patients are affected. The workaround mentioned earlier remains in place until then.