The Netherlands has a plan to fix youth care oversight. One part of it is already solvable.

A few weeks ago, Marlon Domingus, data protection officer in residence at Erasmus University Rotterdam, published an essay on coercive control and Dutch youth care, "De onmacht die zich herhaalt" ("the powerlessness that repeats itself").

In it, Domingus illustrates the problems that arise as a result of data silos in Dutch youth care, and we are encouraged to see that he names Roseman Labs in his article as a working example of the future he's arguing for: where information sharing can support systemic change, while protecting privacy at the same time.

Connecting incidents vs systems

As a data privacy expert and jury member of the Dutch Privacy Awards, Domingus's voice on these serious matters is one worth tuning into. It motivates me to see leaders like him discussing how Roseman Labs fits in these contexts, and it's even better when the article lands on exactly why Roseman Labs was built.

Domingus's essay explores a recurring problem we see across domains struggling with data sharing. Namely, that systems with a limited view on their data ecosystem tend to focus on treating individual incidents, rather than working on patterns higher up in the chain, which often require wider shared context.

Domingus argues that youth care keeps producing the same findings because it's organised around these single incidents. A system built that way struggles to see the pattern in a family's life, or in its own processes. His proposition? Connect organisations in the chain properly, and the system can act on more information. This creates the opportunity to intervene before the next incident occurs. This shared “chain memory” aligns an individual’s journey across the care organisations they meet, rather than creating a new version in each one.

And exactly this aligned journey is what Roseman Labs makes possible in a privacy-preserving way. Organisations can act on a shared pattern without either one seeing the other's underlying records.

Earlier intervention

The City of Rotterdam uses this to catch developmental delays in toddlers across more than a hundred childcare providers, without any provider handing a child's full file to the municipality. UMC Utrecht uses it so hospitals can study signs of child abuse across institutions, without one hospital seeing another's patient records. Once a pattern becomes visible there is still an obligation to act, but surfacing new insights can mean the right people can get involved, or the right support can be given.

This is the foundation that Domingus describes where organisations can work together from the same picture without any one of them losing control of what they hold. It is not a world of more staff, more privacy invasion or more enforcement, but one where alignment becomes stronger across care organizations.

If you’d like to go more in-depth, let me offer you a bit more detail on Domingus’ analysis:

Organising around the system

Domingus looks at why oversight of the Dutch youth care system keeps returning similar conclusions across different reviews. His point: separate inspection findings on crisis intake, local support teams, and child protection services aren't three unrelated problems, but symptoms of a single structural flaw. A system organised around handling isolated incidents will struggle to notice a pattern, whether that pattern is playing out in a child's life or in the system's own recurring findings.

Where the system disconnects

He draws on governance researcher René Clarijs, who identified two forces keeping the system stuck. First, so many actors (government bodies, private organisations, individual citizens) are now entangled that no single party can actually direct the system as a whole. Second, once an organisation settles into a way of operating, institutional habit, vested interests and existing power structures tend to preserve that approach, regardless of whether a better alternative exists.

Clarijs also concluded that Dutch policy in this area mostly funds intensive intervention after problems appear, with comparatively little invested in the everyday support that could stop problems from developing in the first place. Domingus connects this to coercive control, drawing on researcher Evan Stark's framing of it as an accumulated pattern of restriction and control rather than a single event.

Spotting that kind of pattern depends on different organisations pooling what they each know. He argues that pooling rarely happens, largely because of misplaced caution about GDPR, even though the law permits lawful data sharing under the right conditions (referencing last years’ Privacy Awards).

Better data infrastructure, less gaps

Domingus is careful to note that making coercive control a criminal offence, as the UK and parts of Australia have done, hasn't solved the underlying problem on its own. Both jurisdictions still struggle with inconsistent enforcement, and there's documented concern about abusers using court proceedings themselves as another avenue of control after a relationship ends. His conclusion: better data infrastructure addresses one part of the problem, not the whole system around it.

He ends with three recommendations for the ministries overseeing youth care: clear cross-department authority, a named case lead for high-risk situations, and a lawful mechanism for shared data access. He explicitly offers to help build that third piece. We support that wholeheartedly.