Abstract: PO0948
What Actually Happens at Home? A Data Linkage Study Between ANZDATA Registry and Sharesource
Session Information
- Home Dialysis: Disparities and Modality Choice
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 702 Dialysis: Home Dialysis and Peritoneal Dialysis
Authors
- McDonald, Stephen P., Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- Lincoln, Gabriella, Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- Kandamby, Maneesha, Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- Davies, Christopher E., Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
- Duddington, Michelle, Baxter International Inc, Sydney, New South Wales, Australia
- Hurst, Kylie L., Australia and New Zealand Dialysis and Transplant Registry, Adelaide, South Australia, Australia
Background
In late 2018, the Claria automated peritoneal dialysis (APD) system was introduced into Australia and New Zealand. This device provides APD in a nocturnal setting, and automatically transmits data on the treatment session delivered utilising an inbuilt SIM card to a central database (Sharesource). We linked this with clinical and demographic information from the ANZDATA Registry to allow examination of actual events and delivered treatments and alerts.
Methods
Records from Claria APD treatments delivered in 2019 were linked with corresponding ANZDATA records using probabilistic linkage. The data structure (of repeated sessions per individual) was addressed using hierarchical (random effects) models for analyses.
Results
A total of 1641 people had 314,461 APD treatment sessions recorded over 2019. There were a median 189 treatments recorded per individual. Median age of participants was 62.0 years; median time from first kidney replacement therapy was 379 days.
33% of APD sessions had 4 drain cycles, 39% had 5 cycles and 16% 6 cycles. Mean drain time was 20.9 (SD 7.9) minutes; shorter with greater cycle number.
Recorded treatment events included one cycle bypass in 8% of sessions, two bypassed sessions in 8 % and thee or more in 4% of sessions. At least one manual drain in 3.8% of sessions and 0.5% of sessions were user-terminated. One patient alert was recorded during therapy during 27% of sessions, and more than one in a further 10% of sessions. Of all recorded alerts the most common were “low drain volume”, “Check patient line” and “low UF”, all approximately 25% of alerts. The frequency of alerts did not vary across the day of week, but did vary with age, gender and diabetes (lower among older people, females).
The frequency of treatments reported on a Saturday was only 92% of that on a Tuesday. The frequency of bypass (or alerts) was not related to day of week
Conclusion
The variation in frequency of treatment by the day of week suggests "social" causes --- i.e. a "night off" is more frequent on the weekend. Either some form of treatment event or alert is present in a substantial minority of APD sessions. The relationship of these events with technique survival is important, but not yet known. In time this linked data will allow examination of events recorded during treatment with patient outcomes.
Funding
- Commercial Support – Baxter Healthcare