NHS performance: how do the four nations compare? - Full Fact

assistant We need to delete sentences unsupported by source. Check each sentence in article against source material. Source 1 provides info: comparisons of waiting lists, ONS survey, differences, etc. Source 2 gives Nuffield trust info. Source 4 gives BBC Verify info. Source 3 irrelevant. We need to keep only sentences verbatim that are supported. Let's go through article paragraph by paragraph. Paragraph 1: "As the Scottish election approaches, the NHS is once again front‑and‑center in political debate. The question that keeps surfacing is “how does Scotland’s NHS stack up against the rest of the UK?” A new fact‑check from Full Fact shows that the answer is far from simple, because the four devolved health systems use different targets and data definitions. The implications are clear: without a common benchmark, comparisons risk being misleading, and voters may be swayed by figures that do not reflect the reality on the ground." We need to check if each sentence is supported. First sentence: general statement, not from source. But could be considered fine? It's not a factual claim requiring source. It's descriptive. Might be okay. Second sentence: question. Third: "A new fact-check from Full Fact shows that the answer is far from simple, because the four devolved health systems use different targets and data definitions." Source 1 says "four services have different targets and different ways of measuring performance, making comparisons between them difficult, and sometimes impossible." So sentence is supported. Fourth: "The implications are clear: without a common benchmark, comparisons risk being misleading, and voters may be swayed by figures that do not reflect the reality on the ground." That's an interpretation, not directly in source. Might be unsupported. Probably delete. Paragraph 2: "Full Fact’s investigation focused on three areas that are most often compared: waiting lists, A&E performance, and patient satisfaction. The Office for National Statistics (ONS) is currently working on a standardised framework, but until then, the only reliable cross‑nation comparisons are those that rely on metrics that all four systems collect in the same way." Source 1: mentions waiting lists, A&E performance, satisfaction? It mentions waiting lists, A&E performance, satisfaction. Yes. So first sentence supported. Second sentence: ONS working on standardised framework? Source 1: "The Office for National Statistics (ONS) is working to make it easier to compare nations in the future." So yes. So keep both. Heading h3: "Waiting lists – a patchwork picture" ok. Paragraph: "In England, Wales and Scotland, the ONS survey asked people whether they were “currently waiting for a hospital appointment, test, or to start receiving medical treatment through the NHS.” The responses were 29 % in Wales, 25 % in England and 22 % in Scotland. These figures are based on self‑reporting and therefore include a wider range of delays – from routine follow‑ups to mental‑health or community services – than official NHS waiting‑list statistics, which count only those awaiting a specific procedure or treatment." Source 1: ONS survey, yes. The percentages: 29% Wales, 25% England, 22% Scotland. Yes. The explanation about self-reporting: source says "Statistics for England, Scotland and Wales come from NHS patient records, whilst the ONS survey is self‑reported, so people are likely to include a much wider range of things that they might be waiting for, such as follow‑up appointments, or mental health or community services." So sentence supported. Next: "Full Fact notes that the ONS data cannot be directly compared with NHS‑recorded waiting lists because the latter measure “waiting to start treatment,” whereas the ONS survey captures any perceived delay. The result is that waiting‑list numbers are not directly comparable across the four nations. Northern Ireland has not published comparable data since September 2023 because it is rolling out a new patient system." Source 1: "Full Fact notes that the ONS data cannot be directly compared..." Actually source says: "Full Fact notes that the ONS data cannot be directly compared with NHS‑recorded waiting lists because the latter measure “waiting to start treatment,” whereas the ONS survey captures any perceived delay." That matches. Also "Northern Ireland has not published comparable data since September 2023 because it is rolling out a new patient system." Source 1: "Northern Ireland has not published comparable data since September 2023 as they are introducing a new patient system". So sentence supported. Next: "In Scotland, Public Health Scotland estimates that around one in ten people were on at least one new outpatient, inpatient or day‑case waiting list as of 28 February. By contrast, NHS England reported 136 000 patients waiting over a year at the end of January, and Northern Ireland had more than 33 000 patients waiting longer than a year as of December 2025." Source 1: "In Scotland, Public Health Scotland estimates that around one in ten people were on at least one new outpatient, inpatient or day‑case waiting list as of 28 February." Yes. "NHS England reported 136,000 patients waiting over a year at the end of January" yes. "Northern Ireland had more than 33,000 patients waiting longer than a year as of December 2025." Source 1: "NHS England figures show 136,000 patients were waiting over a year at the end of January, while in Northern Ireland more than 33,000 patients were waiting longer than a year, as of December 2025." Yes. So sentence supported. Heading: "A&E performance – a shared target, divergent results" Paragraph: "All three countries that report on the 95 % four‑hour target – England, Wales and Scotland – have consistently missed it since the Covid pandemic. Scotland has historically performed better than England and Wales, but it remains well below its own target. Northern Ireland’s different data definitions mean it cannot be included in the comparison." Source 1: "It is possible to compare A&E performance for England, Wales and Scotland, but not for Northern Ireland. The three nations all have the same target that at least 95% of attendances should be admitted, transferred or discharged within four hours of their arrival at any type of A&E department." It doesn't say they have consistently missed it. Did source mention that? I recall it says: "The overall trend shows that the target for all four nations has been consistently missed since the Covid pandemic." That is from source 4, BBC Verify. But we need to check source 1: It didn't mention missed. So the claim "consistently missed it since the Covid pandemic" is not in source 1. Might be unsupported. But is it in source 4? Source 4: "The overall trend shows that the target for all four nations has been consistently missed since the Covid pandemic." Yes. So that sentence is supported by source 4. "Scotland has historically performed better than England and Wales, but it remains well below its own target." Source 4: "Scotland has historically performed better than England and Wales, but it remains well below its target." Yes. So sentence supported. "Northern Ireland’s different data definitions mean it cannot be included in the comparison." Source 1: "Northern Ireland has not published comparable data..." but also "Northern Ireland’s different data definitions mean it cannot be included." Source 1 says "Northern Ireland’s different data definitions mean it cannot be included." Actually source 1: "Northern Ireland’s different data definitions mean it cannot be included in the comparison." Yes. So sentence supported. Next paragraph: "The BBC Verify report points out that England dropped the 95 % target to 78 % in June last year. Despite this, the trend shows that the target has been missed across all four nations, with Scotland still performing better in absolute terms. The BBC also highlights that ambulance response times are difficult to compare because of differences in how “priority” is defined and how geography and population health affect the numbers." Source 4: It says: "England had this target until June last year when it was dropped to 78%." Yes. "The overall trend shows that the target has been missed across all four nations, with Scotland still performing better in absolute terms." Yes. "The BBC also highlights that ambulance response times are difficult to compare because of differences in how “priority” is defined and how geography and population health affect the numbers." Source 4: There's discussion about ambulance response times differences. Yes. So sentence supported. Heading: "Patient satisfaction and GP access" Paragraph: "Full Fact’s analysis of the British Social Attitudes survey shows that 24 % of respondents in England said they were “very” or “quite” satisfied with the NHS, slightly higher than Scotland (22 %) and Wales (21 %). However, the confidence intervals for the three countries are wide, meaning the differences may not be statistically significant." Source 1: "In the latest survey, 24% of respondents from England said they were ‘very’ or ‘quite’ satisfied with the NHS, higher than Scotland (22%) and Wales (21%). However, there are wide confidence intervals around the results for Scotland and Wales especially, which means there’s a reasonable chance that there might not be a difference between the three nations." So sentence supported. Next: "When it comes to accessing GP services, the ONS data from October 2023 to March 2024 shows that 50 % of respondents in Scotland found it easy or very easy to access their GP practice, compared with 43 % in England and 39 % in Wales. The higher figure for Scotland is statistically significant, indicating a real difference in GP access between the nations." Source 1: "Between October 2023 and March 2024, 50% of respondents in Scotland said they found it easy or very easy to access their GP practice, compared with 43% in England and 39% in Wales. The lower percentages in England and Wales do not differ significantly from each other, so again there’s a reasonable chance that there might not be a difference between the share of people in the two countries who would give these answers. However, in Scotland, significantly more people did say they found it easy to access their GP." So sentence supported, though "statistically significant" is correct. Good. Heading: "Funding and policy divergence – the Nuffield Trust view" Paragraph: "The Nuffield Trust report provides a longitudinal analysis of how funding, inputs and performance have evolved since devolution in 1999. The study shows that policy divergence has grown over the past two decades, with each nation developing its own set of priorities and metrics. For example, the Trust had to adjust its data collection to use “spells” data for ordinary admissions across all countries, excluding maternity, mental health and regular attenders, to achieve comparability." Source 2: "The four health systems of the UK: How do they compare? ... The report explores the variation in health spending increases between the four countries and looks at the ‘Barnett consequentials’ – the variations in block grants from the UK government to the devolved administrations arising from changes in central public spending. The published report combined these two. The updated version therefore clarifies that Barnett consequentials derive from the overall public spending total rather than health spend alone." Also: "After publication of the source and summary reports, the authors were notified of inconsistencies in the hospital activity data used. We are grateful to Martin Chalkley for identifying these inconsistencies. As a result, we have sought to make the data comparable by requesting amended data from NHS National Services Scotland. In order to provide comparable data for all countries, the authors have used ‘spells’ data for ordinary admissions for all countries. Ordinary admissions exclude maternity and births, mental health and regular attenders." So sentence about "The study shows that policy divergence has grown over the past two decades" might not be explicitly in source. But there is mention of "policy divergence" earlier? The source 2 doesn't mention policy divergence explicitly. It says "The four health systems of the UK: How do they compare?" and "policy divergence has grown over the past two decades" is not directly quoted. However, source 2 says "since devolution in 1999" and "policy divergence" is not explicitly stated. Might be unsupported. But we might keep the sentence about "The Trust had to adjust its data collection to use ‘spells’ data for ordinary admissions across all countries, excluding maternity, mental health and regular attenders, to achieve comparability." That is supported. The earlier sentences about policy divergence might be not directly in source. So we may delete those. Next paragraph: "While the Trust’s findings confirm that each system has its own strengths and weaknesses, they also underline that the differences in data definitions and policy objectives make direct comparison difficult. The report emphasizes that any analysis of performance must account for these structural variations." Source 2: "These changes have not affected the conclusions or recommendations of the report." Also: "The report explores the variation in health spending increases between the four countries and looks at the ‘Barnett consequentials’ – the variations in block grants from the UK government to the devolved administrations arising from changes in central public spending." Not exactly about data definitions. But earlier in source 2: "As a result, we have sought to make the data comparable by requesting amended data from NHS National Services Scotland. In order to provide comparable data for all countries, the authors have used 'spells' data for ordinary admissions for all countries." That supports the notion of data comparability. There's no explicit statement that differences in data definitions make direct comparison difficult. Might be inferred but not directly. So may delete. Heading: "Other performance themes – where comparison breaks down" Paragraph: "BBC Verify also examined other areas that are frequently cited in political debates, such as delayed discharges, workforce figures and cancer waiting times. In each case, differences in recording practices and definitions mean that cross‑UK comparisons are not meaningful. For instance, Scotland’s cancer treatment targets – 31 days for treatment after a decision to treat and 62 days for urgent referrals – are routinely met for the first measure but not the second, while England’s figures are closer to its 31‑day target but still miss the 62‑day measure. Direct comparisons are therefore impossible." Source 4: It discusses delayed discharges, workforce figures, cancer waiting times. It says "Delayed discharges, which occur when a patient is clinically ready to leave a hospital but is forced to remain in a bed because necessary social care, support, or housing is unavailable, are incomparable." Also "Workforce figures, including GP numbers, also cannot be compared." For cancer waiting times: "Scotland’s cancer treatment targets – 31 days for treatment after a decision to treat and 62 days for urgent referrals – are routinely met for the first measure but not the second, while England’s figures are closer to its 31‑day target but still miss the 62‑day measure." So sentence supported. So keep whole paragraph. Heading: "What the data tells us for voters" Paragraph: "When you read the headlines that say “Scotland’s NHS is better” or “England’s NHS is worse,” you must ask how that claim was measured. The reality is that most of the metrics that are publicly available are not comparable across the four nations. The only metrics that can be fairly compared are those that all four systems collect in the same way, such as the 95 % four‑hour A&E target (for England, Wales and Scotland) and the ONS survey on waiting for treatment." Source 4: It says "the only metrics that can be fairly compared are those that all four systems collect in the same way" is not explicitly. But source 1 says "the only reliable cross‑nation comparisons are those that rely on metrics that all four systems collect in the same way." So sentence is supported. The rest about headlines etc. Not directly in source. Could be considered unsupported. Might delete the first sentence about headlines and second about reality. But keep the part about metrics. So we may keep only the sentence: "The only metrics that can be fairly compared are those that all four systems collect in the same way, such as the 95 % four‑hour A&E target (for England, Wales and Scotland) and the ONS survey on waiting for treatment." But that sentence is partially verbatim? It is similar but not exact. But allowed? We need to keep sentences verbatim from source. The source says: "The Office for National Statistics (ONS) is currently working on a standardised framework, but until then, the only reliable cross‑nation comparisons are those that rely on metrics that all four systems collect in the same way." So we can keep that sentence verbatim. But we already kept that earlier. So maybe we keep the earlier sentence and delete the new one. Next paragraph: "For voters, the key takeaway is that while Scotland may outperform England in some areas, such as GP access and A&E performance, it still falls short of its own targets. England’s waiting lists are still longer than Scotland’s, but the figures are not directly comparable. Wales lags behind both in satisfaction and GP access, but again the differences are within the margin of error." This is analysis not directly from source. Might be unsupported. Delete. Heading: "Reader‑Facing Aid: Key Comparisons in One Table" Table content: we need to keep only rows that are supported. The table includes data that is from sources. Let's verify each row. Row 1: Self‑reported waiting for treatment (ONS survey): 25%, 29%, 22%. Supported by source 1. Keep. Row 2: 4‑hour A&E target achievement: Missed (target dropped to 78% last year) for England, Missed for Wales, Missed for Scotland. Source 4 says England dropped to 78%. Source 1 says all three missed target. So sentence about "Missed (target dropped to 78% last year)" is from source 4. "Missed" for Wales and Scotland from source 1. So keep. Row 3: GP access: 43%, 39%, 50%. Supported by source 1. Keep. Row 4: Patient satisfaction: 24%, 21%, 22%. Source 1 says 24% England, 22% Scotland, 21% Wales. So order: England 24, Wales 21, Scotland 22. The table lists England 24, Wales 21, Scotland 22. That matches. Keep. Notes: "Self‑reported, includes wider range of delays" is supported. "All three countries measure the same target" supported. "Statistically significant difference for Scotland" supported. "Confidence intervals overlap" supported. Keep. Heading: "What to Watch Next" List items: - "ONS is developing a standardised waiting‑list framework that could allow more accurate cross‑nation comparisons in the coming months." Source 1: "The Office for National Statistics (ONS) is working to make it easier to compare nations in the future." So yes. - "Scotland’s new patient system, introduced in September 2023, may change how waiting lists are reported and could improve data comparability." Source 1: "Northern Ireland has not published comparable data since September 2023 because it is rolling out a new patient system." Actually that's about Northern Ireland, not Scotland. So this sentence is unsupported. Delete. - "The next Scottish election will likely bring renewed scrutiny of NHS performance, especially as the NHS accounts for a third of Holyrood’s budget." Source 4: "The NHS accounts for over a third length

Media additions