Determination model for recourse

Employee Insurance Agency

Shows quickly whether we can reduce costs with a liability claim.

Last change on 5th of February 2024, at 12:39 (CET) | Publication Standard 1.0

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

General information

Theme

Social Security

Begin date

Field not filled in.

Contact information

https://uwv.nl/nl/service-en-contact/contact-uwv

Link to source registration

https://www.uwv.nl/nl/over-uwv/organisatie/algoritmeregister-uwv/vaststellingsmodel-regres

Responsible use

Goal and impact

If you can no longer work due to an accident, you are often entitled to benefits from the UWV. Sometimes we can hold another person liable for the cost of this benefit. For example, if the accident was caused by another person. To quickly see whether this can reduce costs, we use the 'Determination model of recourse' algorithm.

Considerations

If your incapacity for work was caused by another party, we will investigate whether we can have your benefits paid by that other party's insurer. This is called recourse. In this way, we try to reduce the cost of benefits, and thus society.

It also costs money to carry out such an investigation. That's why we look at whether this can save costs beforehand.

Human intervention

Our staff monitors the use of the determination model in the following ways:

• A data analyst puts the data from our systems into the algorithm.
• A staff member randomly reviews the outcomes.

Risk management

The determination model treats every possible recourse case in the same way:

• The algorithm always uses the same data from everyone.
• The algorithm processes all data in the same way.
• The outcome of the algorithm does not affect your benefit or provision.

We ensure that we remain compliant with information security and privacy requirements. We do this in the following ways:

• We work continuously to maintain the algorithm.
• We have carried out a Data Protection Impact Assessment (GEB). An GEB identifies what personal data is used, in what way and whether the privacy of data subjects is adequately protected. The GEB proposes measures to mitigate privacy risks.
• We comply with the Archives Act by deleting data that is not (or no longer) important.
• The algorithm replaces some human actions. This reduces errors and allows us to work faster.

Operations

Data

We use data from clients who have indicated that their disability was caused by another person. This information is known to our Social Medical Affairs (SMZ) and Benefits divisions.

It involves the following data:

• your citizen service number (BSN)
• the date on which you reported sick
• the date on which you recovered
• the reason why you reported sick
• the probable date on which you suffered health damage from a third party
• the type of benefit you need as a result: ZW, WIA, WAJONG, WAO or WAZ
• the starting date of your benefit
• what amounts have been paid to you
• your recourse number

We also use information on the (average) costs of a recourse case. We determine these every year.

Technical design

The algorithm compares the amount we could claim from the liable party with the average execution costs of a recourse case. Cases with lower amounts than the average costs we do not handle.