Public services that work by design: a proposal for Spain's health system

At the launch of Democracia 21 in Madrid on 10 October, Miriam González asked for public services that are well designed and stop leaning on the goodwill of doctors, nurses, teachers and police officers. We set out the rest of her programme in our overview of the party's proposals. Coverage of the speech, such as Press Digital's report, puts it next to her promise to ease taxes while restoring good public services. The phrase deserves a full proposal, and health is the place to start, because the official figures show exactly what she describes.

Trusted staff, a waiting system

The Ministry of Health and the CIS asked 2,602 people about the health service in March 2026, for the first wave of this year's Barómetro Sanitario. 84% of those who used public primary care rated it positively. Asked how much confidence their nurse and their doctor inspire, they gave 8.16 and 7.98 out of 10.

Then comes the appointment. Only 21.8% saw their family doctor the same day or the next. 71.2% had to wait more than a day because nothing earlier was free, and for them the average wait was 10.27 days. When people judge waiting times, specialist appointments come off worst: 69.4% call those waits barely or not at all acceptable, and 62.4% say the same of diagnostic tests. 38.3% think the waiting-list problem has got worse, and 10% think it has improved.

Hospital figures from the national waiting-list system tell the same story. At 31 December 2025, 853,509 people were waiting for planned surgery, 0.8% more than a year earlier. The average wait was 121 days, five days shorter, and 21.6% had been on the list for more than six months. Hip replacements averaged 98 days and knee replacements 105, and 13.7% and 15.9% of those patients had passed the 180-day maximum the law guarantees. For a first specialist appointment, 84.42 people in every 1,000 were on a list, waiting 102 days on average, and 61.5% had been given a date more than 60 days away.

Put the two halves together and you have González's point in numbers. Patients trust the people and wait on the system. A health centre where a doctor stays late to fit in one more patient is running on goodwill, and goodwill wears out: in summer, in rural posts nobody applies for, in the week a colleague is off sick.

The government knows this. The Interterritorial Council approved a Primary and Community Care Action Plan for 2025–2027 in December 2024. It creates a health administrative role to take administrative tasks off clinicians, asks each region for a staffing plan to keep and attract professionals, and promises a ring-fenced budget for the regions and a set of indicators. The Ministry's announcement gives no amount for that budget and no target for appointment times. A good plan, then, still waiting for a design.

What design means here

Jennifer Pahlka, whose Recoding America we explained in an earlier piece, describes how government usually builds things. Instructions flow down from the top, as in waterfall software development, and grow more rigid at every step; rules pile up over successive reforms; the people who deliver the service see data as a tool for blame. Her remedy, as Rebecca Tushnet's review sums it up, gives delivery teams the authority to adjust as they go and has oversight ask what user tests are showing now.

James C. Scott, explained here, adds the warning. Schemes designed from the centre fail when they flatten complex arrangements that planners do not understand, and Yale's description of his book insists that local, practical knowledge counts as much as formal knowledge. In a health centre, that knowledge belongs to the nurse who knows which patients will call three times before they come in, and to the receptionist who knows which form is pointless.

There is a European example of what happens when you trust that knowledge. Buurtzorg, the Dutch home-care organisation, runs on self-managing teams of nurses. When researchers at St George's, University of London studied an adapted version in an English district nursing service in 2018, they found it feasible and acceptable to patients, carers and GPs; patients valued the continuity, and nurses found it a more attractive place to work. The same study could not put a figure on costs and found the teams struggled to be recognised inside a large, bureaucratic organisation. That last finding is the design problem in one line.

The proposal

The left's case comes first, because without it nothing else works. Spain's health service stays universal and publicly funded, and primary care gets the ring-fenced money the 2025–2027 plan promises, with the figure written into each year's budget and published. Posts that nobody fills in rural areas get the pay and conditions it takes to fill them.

The market-liberal case comes next. Every health centre gets real control over how it organises its work: its schedules, its mix of doctors, nurses and administrative staff, and part of its budget. In return, each centre publishes every month the days to the next free appointment, the share of patients seen within 48 hours and how long its patients wait once referred to a specialist, on a public page any patient can read and compare. Regions that let patients choose their centre or their doctor make that choice meaningful by publishing those figures.

The piece that belongs to neither camp is how the redesign happens. Each regional health service sets up a small delivery team, Pahlka-style, which spends weeks inside volunteer health centres mapping every step a patient takes and every form a clinician fills in. The centre's own staff lead the work and decide which steps to drop, merge or hand to the new administrative staff, because they hold the knowledge Scott says plans miss. Every change is tested in a few centres, measured on those public figures, and spread when it works. And the regional health ministries accept a burden budget: any new instruction, form or circular comes with an estimate of the clinical minutes it costs per consultation, and goes through only if an equal cost is removed elsewhere.

Done this way, a nurse's goodwill goes into patients again, the time clinicians spend on paperwork shows up in a number someone has to answer for, and the design improves every year because the people who know where it fails are the ones redesigning it. The same method can serve schools and police stations next.