Part 2 of The State of DHCW. DHCW's own January 2026 evidence concedes that LIMS, RISP, Intensive Care and Connecting Care 'have not delivered to their original timetable' and that strengthened governance since 2023 'has not resolved issues around the pace of delivery.' The programme ledger documents thirteen programmes: WCCIS failed and at end of life, LIMS's contract terminated, RISP Amber-Red, WICIS paused by ministers, maternity procurement terminated, and the integration layer everything depended on still in private beta after three years of commitments.
Part 2 of The State of DHCW. Numbered citations resolve at the sources page.
The most damaging single piece of evidence is not external. It is DHCW’s own written evidence to its Public Accountability Meeting with Welsh Ministers, submitted on 15 January 2026:
“As described above, several large collaborative digital programmes have not delivered to their original timetable, including: LIMS, RISP, Intensive Care and Connecting Care. Concerns around the delivery of major programmes has resulted in DHCW’s escalation. DHCW has strengthened programme governance, assurance, transparency and oversight since 2023, but this has not resolved issues around the pace of delivery … However, programmes continue to experience delays to delivery timetables.”[4]
That is an admission that four years of governance reform did not change delivery outcomes. The programme-level detail below is drawn from Audit Wales, DHCW’s own board and committee papers, health board committee reports and the Senedd record.
The programme ledger
| Programme | Status | Documented record |
|---|---|---|
| WCCIS Welsh Community Care Information System | Failed; end of life | The 2015 business case assumed all 22 local authorities and 7 health boards live by December 2018. The Auditor General found in October 2020 that "implementation and roll-out of WCCIS are taking much longer and proving more costly than expected … the prospects for full take-up and benefits realisation remain uncertain," that the Welsh Government's four February 2020 deadlines were missed ("None of the deadlines were met"), and that the contract framework "lacks liquidated damages clauses for delivery delays."[17] A June 2021 independent user survey found the system "was having a more negative than positive impact on most users' ability to do their work."[16] Aneurin Bevan UHB missed go-live dates in June 2019, November 2021 and March 2022; Betsi Cadwaladr missed April 2017 and then November 2021.[16] By 2022 the Auditor General judged the programme "at a critical juncture," requiring "strong leadership … to ensure value for money from the £30 million investment to date," and recorded that at least two local authorities were seeking to exit.[16] DHCW told the Senedd in October 2022 that quantifying the benefits was "quite a tricky question to answer."[22] By May 2025 DHCW's own Programmes Delivery Committee recorded: "There is currently no funding for ongoing support of WCCIS," with Connecting Care named as its replacement.[36] |
| LIMS 2.0 All-Wales laboratory system | Terminated contract; delayed | Procurement began in 2021. Citadel Health was contracted to deliver by September 2023 (a later health board account gives July 2025 as the contracted date); "the contract was terminated during FY2022/23," and the incumbent supplier had to agree a five-year extension.[51] DHCW's own evidence: "A supplier contract was awarded in 2021, with a plan to implement by September 2023. This timetable was not achieved, which led to a programme transfer to DHCW and a new supplier contract."[4] Hywel Dda UHB reported in July 2025 that "key milestones have been delayed, and it is now evident that the overall programme timelines will not be met," with a further £1.6m being sought, and warned that "delays … may extend beyond the end-of-life date of the current system. Should this occur, it could lead to a complete failure of the pathology service."[51] DHCW re-escalated a corporate risk for WLIMS 2.0 delay in July 2025, citing a compressed timetable placing "significant pressure on the teams involved."[37] As at January 2026, two of five tranches — Blood Sciences and Blood Transfusion — had completion dates "under review."[4] |
| RISP Radiology / PACS replacement | Delayed; Amber-Red | DHCW's Q4 2023–24 portfolio report: "The Radiology Informatics System Procurement (RISP) has pushed out go-live dates for most health boards."[45] DHCW's January 2026 evidence records four organisations live in 2025, "the programme timetable has come under increasing pressure," and several organisations requesting later dates. Critically, it identifies a structural procurement flaw: the master services agreement "allows organisations to unilaterally move their go-live date, which has an impact on other organisations because the series of implementations is already tightly sequenced."[4] DHCW's February 2026 Programmes Delivery Committee recorded overall RAG status "Amber-Red" and that "managing separate contracts has proven difficult."[35] DHCW also concedes that running LIMS and RISP simultaneously "was not the original intention."[4] |
| WICIS Welsh Intensive Care Information System | Paused by ministers | DHCW's May 2025 committee minutes: "DHCW is now working with Welsh Government and NHS Executive colleagues to determine the next steps, following Welsh Government instruction to pause the programme, pending an independent review."[36] The March 2025 escalation framework required DHCW to undertake "an assessment on the next steps for this programme, and how it should be resourced and led."[2] The April 2026 framework cites WICIS repeatedly as an example of "repeated reprofiling of milestones … without materially different outcomes" and of "delivery and readiness concerns escalating late."[1] |
| Digital Eyecare Ophthalmology EPR and referrals | Five years late | Launched 2021 under Cardiff and Vale UHB with £4.8m for a cloud-based referral and monitoring system; transferred to DHCW on 1 June 2023.[54] A Gateway 0 review concluded: "Successful delivery of the programme is in doubt with major risks or issues apparent in a number of key areas," and Welsh Government imposed a due-diligence pause.[31] Implementation-readiness work found "tactical implementation would take between 8 months to 2.5 years," with health boards unable to commit without funding.[31] A Senedd opposition motion of 27 September 2024 formally regretted that "the electronic patient record and referral system, first launched in 2021, is still not operational across Wales."[28] At committee on 14 May 2025, chair Peter Fox MS said lack of digital infrastructure was "holding back" eye care improvement; Cardiff and Vale's digital director said full national rollout was hoped for in early 2027.[54] By January 2026 DHCW was procuring a separate "tactical solution for Optometry e-referrals" commissioned in June 2025 — a parallel workaround to its own national programme.[4] |
| NHS Wales App Digital Services for Patients and Public | Partial; behind England | Over 650,000 registered users, about one in five Welsh adults, with roughly a million logins a month.[4] But Senedd Research notes "the app still offers fewer features than its English equivalent, a gap that was highlighted during scrutiny in the Sixth Senedd."[26] DHCW's February 2026 committee papers record that "the NHS Wales App remains partially delivered, not yet live in Cardiff and Vale, though it is live in six of seven health boards."[35] The independent Ministerial Advisory Group on NHS Wales Performance and Productivity recommended in April 2025 that "no health board should move forward with any EMR or App development until the roadmap is agreed," and that "full consideration should be given to aligning the NHS Wales App with the NHS England App."[9] The April 2026 framework cites the App as an example both of "limited evidence of benefits realised at scale despite sustained investment" and of milestones achieved "without assured local capacity, readiness or benefit realisation."[1] |
| Digital medicines EPS · ePMA · Shared Medicines Record | Very late start, then good pace | England and Scotland introduced electronic prescribing in 2009; Wales issued its first electronic prescription in November 2023 — a 14-year lag.[26] In 2022 Community Pharmacy Wales and the Welsh Pharmaceutical Committee described the pace as "incredibly slow," at a point when "prescribing remains predominantly paper based in Wales."[27] The March 2024 milestone for a persisted Shared Medicines Record was missed and reprofiled to June 2024.[45] Hywel Dda reported a ten-week ePMA delay in October 2025 "due to significant integration and configuration challenges," pushing its target from January to April 2026.[52] Against that, EPS rollout is now "on track to complete rollout across Wales by November 2026, a year ahead of the original programme timetable," with more than 15 million items dispensed by February 2026.[4][26] |
| Integration Hub replacement for Fiorano | Private beta after 3 years; unfunded | Committed in the IMTPs for 2023–26, 2024–27 and 2025–28 in near-identical terms.[49][79] As at 31 December 2025 it was in private beta with no flows in production, first flow targeted January 2026; internal staff onboarding "has not been possible and onboarding dates re-forecast"; API use "yet to be defined"; health board testing flagged as a potential blocker.[77] Recorded as not funded in the February 2026 Board finance report.[63] The legacy platform contract now ends June 2027, with additional cost exposure "if migration slips."[72] Full analysis below. |
| National Data Resource | Ten-year programme; majority of milestones missed | DHCW internal audit, April 2024: phase 3 funding "was reduced in year by 27% … from £8.2m to £6m. This reduction has had an impact on the ability of the programme to deliver all milestones," with benefits pushed into a later phase.[39] NDR was one of the three programmes DHCW itself named as driving the March 2025 escalation, alongside LIMS and RISP.[4] DHCW's own December 2025 assurance report: "Of the 13 Programme milestones due at the end of Q3/Dec 25, 6 have been achieved, 7 have been reforecast or change controlled," with resources "diverted to urgent priorities," interim governance still in place and "unprecedented resource constraints."[77] By May 2026 NDR was among the causes of Mission 1 falling from Amber to Amber/Red.[72] Full analysis below. |
| GP systems INPS/VISION migration | Supplier collapse absorbed | In 2024–25 In Practice Systems announced exit from the Welsh market, triggering migration of 196 practices; in December 2024 it entered voluntary administration "creating a significant risk for the 154 practices yet to migrate."[4] DHCW secured continuity with a new supplier and accelerated migration, but the episode also paused DHCW's own internal operating-model rollout in its Primary, Community and Mental Health Directorate.[12] A second supplier (Audit+) announced withdrawal in April 2024 with "no comparative solution found in the marketplace."[4] This is fairly characterised as competent crisis management of an externally caused problem. |
| Connecting Care | Late-starting successor | Established in November 2023 to replace WCCIS as it reaches end of life, and recorded by Audit Wales at that point as "currently unfunded."[13] The March 2025 escalation framework required delivery "to the agreed scope and business case, once agreed" — i.e. no agreed business case existed at escalation.[2] Named by DHCW in January 2026 among programmes that "have not delivered to their original timetable."[4] |
| Digital Maternity Cymru | Procurement terminated; programme closed | Established 2023 after a 2022 discovery phase to procure an all-Wales digital maternity system and app. National procurement commenced October 2024; first-phase evaluation completed December 2024; "the outcome was a decision to terminate the procurement process."[71] Supplier feedback was that "the service management requirements were disproportionate compared with what the market could realistically provide," and the lessons-learned review recommended securing "formal, organisation-wide buy-in at the outset."[71] DHCW's annual report confirms the outcome: "Welsh Government is now funding individual health boards to implement locally. Therefore, the national programme, managed by DHCW, will close in 2025-26," with only a data-standards project retained.[67] Named in the April 2026 Level 4 framework among programmes that "weakened confidence that nationally announced milestones can be delivered to plan."[1] |
| Single patient record | Not delivered | The electronic patient record has been Welsh policy since the 2003 Informing Healthcare strategy.[21] In March 2025 the First Minister told the Senedd the Welsh Government was still "developing a national business case for the next generation electronic health record system in Wales."[30] In June 2026 the incoming Cabinet Minister for Health and Care said the new government had used its "first 100 days … to reset the approach to digitisation … including a road map of a single integrated patient record for Wales."[29] Twenty-three years after the policy was set, the deliverable remains a roadmap. |
The integration layer: the dependency everything else waits on
Every programme in the table above depends on the same thing: the ability to move data reliably between systems. This is the layer where DHCW’s difficulties are most consequential and least visible, because integration failure does not produce a headline — it produces other programmes slipping. DHCW’s own documents are unusually explicit about it.
The size of the problem, on DHCW’s own numbers
The National Target Architecture project’s current-state mapping quantified what had previously been assumption. It “exposed the number of separate digital systems in use (over 1400), how the patient record is fragmented and duplicated (across 300 separate systems) and helped to identify legacy systems which are at or near end of life.”[4] DHCW describes the prize in its own terms: less fragmentation “will reduce the time needed for technical and user testing and will reduce the running cost of ‘integration overhead’.”[4] Audit Wales found Betsi Cadwaladr alone “operates over 400 digital systems across its estate, creating complexity and integration challenges.”[82]
The Integration Hub: promised in three successive plans, one flow live in five years
The replacement integration platform has been a stated DHCW commitment since its first medium-term plan, in near-identical language each time:
| Plan | Commitment |
|---|---|
| IMTP 2023–26 | "We will build a data and integration hub to allow data to move around securely and safely. This means data is available across organisational and geographical boundaries to the right person at the right time to enable better care."[49] |
| IMTP 2024–27 | "We will establish product roadmaps, such as a new integration hub and further adopt open standards."[79] |
| Q3 2024–25 portfolio report | "We will build a data and integration hub to allow data to move securely and safely across organisational and geographical boundaries."[79] |
| IMTP 2025–28 | "Integration Hub" listed among named deliverables alongside Care Data Repository, Target Architecture and Shared Medicines Record.[79] |
DHCW’s own Major Programmes Assurance Report for December 2025 — which rates every major programme’s “delivery confidence” — shows where that had actually reached after nearly three years of commitments. As at 31 December 2025 the Integration Hub had no flows in production at all: “basic set up and flow in Azure environment achieved in Q1 2025. Beta Delivery phase commenced with goal of first flow to production planned Q4 2025/2026,” with the first prioritised flow (Master Patient Index and Public Health Wales) aiming “to go live with this flow during January, with dependency on assurance and sign off from groups.”[77] The programme was in private beta.
The reported blockers read as a précis of the whole DHCW problem:
- It cannot staff the work itself. Timelines and Resources were both rated amber-green “due to delays in onboarding internal staff to the product due to upskilling required,” and “challenges in the recruitment for key roles required to backfill delivery partner colleagues.” Introductory training completed in Q2, “with plan to onboard to Integration Hub product following this, but has not been possible and onboarding dates re-forecast.”[77] The build depends on a “hybrid team” with an external delivery partner whose funded work package ended in February 2026.[77]
- It is blocked on the adopters. “health board testing could become a blocker”; workshops with health boards were deferred to early 2026 “due to winter pressures across NHS Wales”; and on easing testing pressure, “Meetings on this have taken place, but due to conflicting priorities across NHS Wales, have been unable to move this forward.”[77]
- The API strategy is undefined. “there are also discussions around use of APIs for certain flows, these are yet to be defined, discussions with teams are ongoing.”[77] For an integration platform in beta, this is a striking gap.
- It is not funded. The February 2026 Board finance report records “cyber funding confirmed; e-referrals/integration hub not funding, may need budget rephasing.”[63]
- A hard deadline is now approaching. In December 2025 the “Current Fiorano contract ends in June 2026, with an option to extend.”[77] The option was evidently taken: by May 2026 the Audit and Assurance Committee “were alerted to the contractual end date in June 2027 for the legacy platform of the Integration Hub. If migration slips, additional costs may be incurred.”[72] DHCW therefore has roughly a year to migrate the national messaging estate off a platform it has spent three years planning to replace, starting from one flow.
Fiorano: the messaging fabric it is still running on
The legacy engine is Fiorano, and its record is poor. In 2021 Cwm Taf Morgannwg UHB reported “3 outages in the past 14 days associated with the DHCW Fiorano integration engine which have taken down numerous clinical applications.”[53] As late as August 2024, the substantive work under way was not replacement but a version upgrade: an assurance plan whose scope was “the migration of all workflows from Fiorano version 10 into Fiorano version 13.”[80] The single point through which national clinical messaging passes has thus been sustained by in-place upgrades while its replacement remained in beta.
The National Data Resource: a ten-year programme, and half its milestones missed
The NDR is the data half of the integration problem — and it is enormous in scope. DHCW describes it as a programme that “over ten years … will design, develop, assure and deploy a new cloud data platform,” managing “more than twenty delivery projects,” with around a third of its annual budget “disbursed to federated partner organisations.”[4] Its components include the Care Data Repository (intended to replace the Welsh Care Records Service and Welsh Demographic Service), the National Data and Analytics Platform (replacing the national Data Warehouse), a secure data environment, an API Management service, and a learning platform.[4]
DHCW’s December 2025 assurance report on NDR is candid:
“there are recurring themes of planned work being deferred in order to respond to emerging priorities and continued resource constraints which have become more widespread across the product streams. … Since September programme resources have been diverted to urgent priorities, namely NHS Wales App appointments and referrals, and Maternity system integrations. This urgent additional work has impacted planned delivery. Of the 13 Programme milestones due at the end of Q3/Dec 25, 6 have been achieved, 7 have been reforecast or change controlled. … Interim Programme Governance arrangements remain in place. … During this period the programme has experienced unprecedented resource constraints such as changes to team structures, vacancies, sickness and other absences.”[77]
Set against a ten-year, twenty-project programme, the list of live services actually built on the NDR platform is thin: the Wales Growth Chart Module in the Welsh Clinical Portal, the Shared Medicines Record, hospital appointments in the NHS Wales App, and ADT data feeds to local maternity systems.[4] Phase four — “building the whole digital record particularly through migration of historic data” — was still in progress at January 2026.[4] And the dependency runs the other way too: hospital electronic prescribing waits on NDR, since the Care Data Repository was working on “the migration of Allergies Data into CDR which is required by the Electronic Prescribing Medicines Administration.”[77]
The trajectory is downward. NDR lost 27% of its phase 3 funding in-year in 2023–24, from £8.2m to £6m, which “had an impact on the ability of the programme to deliver all milestones.”[39] It was one of three programmes DHCW itself named as driving the March 2025 escalation.[4] Rated Amber/Green in December 2025,[77] by May 2026 it was among the causes of a whole-mission downgrade: “the change in Mission 1 RAG status from Amber to Amber/Red, driven by challenges within key enabling areas, specifically the National Data Resource (NDR), Cyber, Information Governance, and Cloud.”[72] At the same meeting the committee was “alerted to National Data Resource Programme risks, including funding uncertainty and recruitment pause, which may impact delivery if unresolved.”[72]
Architecture and standards: started late, and stalled
The most telling date in this whole section is the start date of the architecture work. “The National Target Architecture project is commissioned and funded by Welsh Government. The project was established in June 2025”[78] — four years after DHCW was created, and three months after the Level 3 escalation made “national current and future target enterprise architecture development and design” a mandated milestone.[2] DHCW then “procured an architecture repository solution” (Ardoq) and “procured an external consultancy” to collate the technical data, agree the architectural principles and produce the current-state report.[78] In July 2025 the Board was told the target was to have just ten organisations mapped by mid-September.[70]
The Strategic Investment Plan work then ran into the same wall as every other national programme — the adopters:
“The project was unable to gather as much evidence as was intended to inform development of the strategic investment plan. For example, absences of evidence included local contractual timetables to expiry, renewal and procurement, and three-year investment plans detailing committed and proposed investments in new digital platforms, products and services. … If this evidence is not received there is a risk that investment decisions are not fully informed of context and not properly co-ordinated at the system level, which could lead to reduced value for money due to impacts such as duplication, sub-optimal contracting.”[78]
The milestone was nonetheless signed off — but DHCW’s own committee recorded what it actually is: “The plan is intended to inform the next phase and acts as a framework for developing future business cases, rather than being a business case itself.”[72] A framework for future business cases, five years in. And by May 2026 the committee was “alerted to the pause of delivering the National Target Architecture project plan for 2026-27 due to recruitment pause, and funding uncertainty due to the DPIF pause and review.”[72]
On standards, DHCW has the machinery — the NHS Wales Data Dictionary, a Data Quality service, the Welsh Reference Data and Terminology Service, clinical classifications support, and representation on INTEROPen and UK FHIR.[4] The gap is adoption and delivery, not apparatus. National data standards work on WCCIS was explicitly descoped in the 2022 strategic review and considered for transfer elsewhere in DHCW.[16] A Flintshire practice manager told the Senedd in July 2025 that “changing to SNOMED coding has been part of the Welsh Government route-map for seven years, but it still hasn’t happened in primary care.”[32] And the Cabinet Secretary’s February 2026 letter puts the residual dependency plainly: realising the NDR’s value “depends on health boards improving data quality, meeting data standards and accelerating data contribution to the repository.”[3]
What the adopters say
Audit Wales’s 2026 digital transformation reviews of individual health boards provide independent corroboration from the receiving end. At Cwm Taf Morgannwg:
“the Health Board has raised concerns that roles and responsibilities are not always clear between itself, DHCW and Welsh Government, especially when it comes to how national digital programmes are delivered and supported. The Digital and Data Highlight Reports to the Operational Delivery Committee show that delays to national digital programmes are ongoing and significant. These delays make it harder for the Health Board to manage digital risk.”[81]
At Betsi Cadwaladr, Audit Wales found the board “supports national systems and the Once for Wales approach … However, putting these systems into practice remains difficult. Old systems, limited staff readiness and gaps in infrastructure slows progress and reduce the benefits the Health Board can gain from national [systems].”[82]
Assessment. Integration is where DHCW's structural position bites hardest, and the evidence supports the charge of sustained weakness. A replacement integration platform has been promised in every plan since 2023 and had one flow in production five years into the organisation's life; its API approach was still "yet to be defined" in beta; it was unfunded in the 2026 budget; and it now faces a June 2027 legacy contract cliff. The national data platform is a ten-year programme that missed seven of thirteen quarterly milestones, is running on interim governance under "unprecedented resource constraints," and had resource pulled off it to rescue the App. The architecture work that should have preceded all of it began in year five, was largely done by external consultants, could not obtain basic contract and investment data from health boards, and produced a framework for future business cases rather than a plan.
Two qualifications are owed. First, much of this is genuinely not soluble by DHCW alone: an integration platform cannot go live faster than health boards can test it, and an investment plan cannot be built from data the health boards do not supply. Second, the underlying operational service is stable — the Care Data Repository, analytics platform and secure data environment are live, and the Welsh Clinical Portal does share records across every health board boundary, which England has not matched.[26]
But the sequencing failure is DHCW's and the Welsh Government's jointly, and it is the deepest cause in this record. Wales spent 2021–2025 running large national application programmes — WCCIS, LIMS, RISP, WICIS, maternity, eye care — on top of an integration and architecture foundation that had not been built, and in several cases had not been designed. The programmes then failed in precisely the way an absent integration layer predicts: on interfaces, data migration, message handling, testing capacity and dependencies. Building the foundation last is not a delivery problem that better milestone reporting can fix.