The Government says it will give localities more control over healthcare decisions. But how?

New Zealand’s political reflex is that any problem can be resolved by further centralisation. Students will be officially banned from having cell phones at school from Term 2. The decision could have been left to individual schools. Each knows a lot more about local circumstances than the Minister of Education does (or I do). But the New Zealand way is a central directive.

On the other hand, sometimes centralisation is needed. Historically, there has been an ongoing process of consolidation of secondary healthcare. Hence cottage hospitals scattered throughout the country being slowly turned into a nationwide hospital system. Even so, there is a hierarchy among the hospitals. Today a person with a serious heart condition in Nelson – which has as good a provincial hospital as there is – is likely to be flown to Wellington.

Medicine has become more specialised and is evolving rapidly. That suggests that hospital care needs to be built on advanced medical centres attached to a base hospital. At best there is sufficient scale in New Zealand for only five centres offering the specialised levels of care which provincial hospitals are unable to provide: Dunedin, Christchurch, Wellington, Hamilton and Auckland.

Sometimes a single agency makes sense. As much as the Big Pharma are attracted to weak local purchasing, a single Pharmac works for us. Both the interdependence of tertiary and provincial hospitals and the mobility of New Zealanders means the IT configurations among health regions need to be able to talk to each other; currently they can’t.

Even so, the form of centralisation of the system in Health NZ (HNZ) did not seem to make much sense. The most widely used justification for the redisorganisation was the ‘post-code lottery’ – the access to treatment varying by region. The response has been typical of so much policy in New Zealand. A correlation was treated as causation and policy proceeded on the basis that if we abolish regional governance there will be no post-code lottery. No attempt was made to explain the disparity, although it does not take a lot of imagination to think of explanations for the differences which would not lead to a centralisation policy.

Nor does the current centralisation policy remember that while the health redisorganisation of the early 1990s was focused on competition and privatisation, there was also a concern that some areas suffered from a lack of attention from the central hospital. A positive reason for separating Middlemore Hospital from the rest of the Auckland hospital system was that South Aucklanders’ health had been neglected.

The real reasons for policy changes are often different from the stated reasons. A possible reason was that the Ministry of Health was judged to be failing and it was thought better to set up a new agency rather than redisorganise the Ministry. Possibly the shift to national pay-and-condition scales was a consideration in favour of centralisation. The redisorganisation of the early 1990s left industrial relations in the hands of individual CHEs/DHBs. Over time, that decentralisation has been replaced by a system of national awards.

One factor, surely, was that the population-based funding model was failing. It was first introduced in the early 1980s and was, at the time, a progressive attempt to move away from a rigid funding system based on historic proportions. There were later refinements but the formula appears never to have been properly adjusted for the cross-border flows of patients referred by  provincials to tertiary centres, nor for differences in population density and concentration nor for economies of scale.

Crucially, to be equitable the funding formula required that each DHB had a capital structure which generated a similar level of productivity together with the assumption that the shocks each DHB experienced were small. Both assumptions were wrong as vividly illustrated when the Canterbury DHB faced the aftermath of the Canterbury earthquakes and the Mosque Massacres. The new system is not bound by the old funding formula and may be able to refine it, although it may end up making ad hoc decisions responding to perceived short-term pressures.

The latest redisorganisation does not really address these concerns. Rather, HNZ has been charged with designing the new system. One advantage it has is that it is gaining hands-on experience, unlike most top-down organisations charged with redisorganisation. But the pressures of dealing with the minutiae of that experience may divert its attention from the overall picture. Without external pressure it is unlikely to unwind if it discovers it is overcentralised.

But is HNZ hands on enough? It has a hierarchical structure which means that, as is common among generic managers, the leadership does not connect well with the knowledgeable below. HNZ need not listen to the hospitals it runs. The new organisational structure adds at least one further managerial layer into the system.

In an RNZ interview that got overlooked in the rows over scrapping the Māori Health Authority and the existing smoking reduction policies, the new Minister of Health, Shane Reti, says he is shifting more health decision-making back to the regions. ‘There are some parts that need to be owned by the centre, absolutely, but we need to be very careful because what has happened here is we've lost local accountability. We've lost local decision making and it's all owned by the centre.’ He stopped short of saying district health boards would be reintroduced but said IT systems and key services like radiotherapy machines were examples of what should remain centrally managed.

What the minister has in mind is unclear. Giving the local health deliverers more autonomy may sound an excellent idea but how are they to be held to account? The Minister appears to have ruled out elected boards (last introduced by the Clark Labour Government and revoked by the Hipkins Labour Government). I am not particularly sympathetic to such boards, having had friends elected to them who felt they had little influence; those with greatest integrity chose not to stand again.

Nor do I have much sympathy for the fashion of introducing targets. We have had a lot of experience with them in economics summarised in Goodhart’s Law that when one specific goal is set, people will tend to pursue that objective regardless of the consequences. An example was that when emergency departments were given a target time for processing admissions, cases were left in ambulances outside, only being admitted when the target time could be achieved. Never forget Gilling’s Law: ‘the way you score the game shapes the way it is played’.

(The government’s announced five targets are hardly inspiring. Only one addresses primary care and that – ‘95 percent of children to be fully immunised at 24 months of age’ – while worthy, ignores the numerous recommended vaccinations for adults and older children, not to mention a multitude of general practice issues.) The lack of mention of general practice will lead to the equivalent of the needy waiting in ambulances outside. 

Ian Powell’s sober assessment of targets concludes more sympathetically towards the previous Labour Government’s indicator approach than the Key-English one of hard targets.

The problem arises because a health system, like an economy, has multiple objectives which cannot be reduced to a single number in the way a business has a profit concern. (Even that gets corrupted by financial chicanery.) Simplifying the multitude to a single measure – as with the focus on changes in real GDP – is simple, foolish and distorting.

It is unreasonable to expect a new government to have fully formed its views its first hundred days – being in government and doing things is so much harder than being in opposition and criticising. It would be good if this one decides to pursue a culture change in which local generic managers focus on supporting those who deliver health care to individuals. That is a long way from a philosophy of centralisation.