Care profile algorithm

CIZ Centre for Healthcare Indication

The algorithm helps the CIZ employee choose the care profile that best suits the client's care needs. A care profile states what care a person needs.

Last change on 31st of October 2024, at 16:40 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

Health and Healthcare

Begin date

2013-11

Contact information

https://www.ciz.nl/hulp-en-contact#contact

Responsible use

Goal and impact

The algorithm is a tool for the CIZ employee to choose the most appropriate care profile for the client. This involves how much and what care is granted.

Support from the algorithm helps the CIZ to issue the same care profile in similar situations.

Considerations

The algorithm is in use to also assign similar type and amount of care in similar situations.

The algorithm supports the staff making the decision.

This is also a moment of reflection for the employees. They test whether the image they have of the situation and the recorded file correspond to the outcome of the algorithm.

Human intervention

The CIZ employee first determines whether there is access to the Long-Term Care Act (Wlz). Once access is determined, the most appropriate care profile is chosen. The algorithm helps choose which care profile best suits the situation, i.e. how much and what kind of care is appropriate.

The algorithm shows the best-fit care profile and the best-fit alternative, after which the staff member decides whether one of these two proposed care profiles is appropriate or whether the staff member thinks another care profile is a better fit after all.

Risk management

CIZ processes sensitive personal information and is aware of the potential risks involved. Therefore, this algorithm does not use personal data such as name, address or data on ethnicity. Thus, the algorithm only looks at filled data on health status.

The employee makes the decision, not the algorithm. This staff member is familiar with the content of the different care profiles and and, based on his professionalism, has the ability to make a choice other than the care profile suggested by the algorithm.

To assess the performance of the algorithm, the algorithm was reassessed after some time. The algorithm does not learn itself, eliminating the risk of incorrect interpretations by the algorithm.

Further actions to check the algorithm are not planned, as this algorithm will no longer be used in CIZ's new system that will come into operation in 2025.

Legal basis

The Long-Term Care Act (Wlz) describes the criteria for access to care from the Wlz and the role of the CIZ to examine this (indication).

The care profiles are described in the Regeling langdurige zorg (Rlz)

Links to legal bases

• Wet langdurige zorg (Wlz): https://wetten.overheid.nl/BWBR0035917/2024-07-01
• Regeling langdurige Zorg (Rlz): https://wetten.overheid.nl/BWBR0036014/2024-07-01

Operations

Data

The CIZ employee does research and fills in scores for the client's severity of disorders and impairments. The algorithm uses these scores and makes a calculation to determine which care profiles fit best. The algorithm can do this because it has been trained on lots of past decisions that have taught it which care profile best fits a combination of scores.

Technical design

The client's functioning and care needs are identified by the staff member during the indication examination. The findings of the examination are recorded by completing a questionnaire on impairment and (functional) limitations scoring. These questions are grouped into nine domains of self-reliance, and a scale score is calculated on each of these domains. The algorithm calculates from these scale scores, through weighting coefficients and cut-off values, which care profile fits best and which is the best alternative thereafter. Both are shown with their own percentage 'correctness'.

The weighting coefficients and cut-off values come from a regression analysis performed on historical data. There is no automatic adjustment of these values based on new data. Thus, the algorithm is not self-learning. This means that the CIZ can always explain on the basis of which logic the algorithm arrived at the advice.

External provider

Developed internally in collaboration with Equalis(sirm)