Performance indicators in the NHS: are they pathdependent

1
Performance indicators in the NHS are they
path-dependent?

• Steve Harrison (Manchester)
• George Dowswell (Manchester)
• Christopher Pollitt (Leuven)
• Roland Bal (Erasmus)
• Sonja Jerak-Zuiderent (Erasmus)
• FUNDED BY THE UK ECONOMIC SOCIAL RESEARCH
  COUNCIL PUBLIC SERVICE PROGRAMME

2
PIs in the English NHS the story -1

• A new sense of rationalism in public sector,
  manifested by introduction in 1960s of such
  techniques as PPBS, PAR (white paper 1970),
  planning, MbO, statistics, work study etc, all
  aiming to relate programme outputs to inputs.
  Interest waned by mid-1970s.
• Royal Commission on NHS (1976-79) recommended
  better information for NHS (as previously has
  Guillebaud Cttee of 1956) led to establishment
  of Korner Committee 1980, reported 1982.
• Conservative govt FMI announced 1982 in response
  to parlimentary criticism focus on output
  measurement.
• The invention/ discovery of specifically NHS PIs
  1982 as element of regional review process.
  Opportunistic re-labelling of existing routine
  data plus borrowed academic work. Underlying
  theory seems to have had elements of formative
  evaluation. PIs - assumed to be imperfect proxy
  for performance (1983 package 70, 1985 450,
  1989 2,500, 1994 400)

3
PIs in the English NHS the story - 2

• The assembly of PIs into comparative league
  tables of institutions, (necessarily?)
  accompanied by a reduction of indicators
• The making public of the above league tables
  (1994) in form of a guide for patients, using
  hotel-style star ratings. Really aimed at
  applying pressure on providers, rather than
  supporting choice Klein (2001).
• The invention of regulatory institutions, such as
  the former Commission for Health Improvement,
  whose functions include the making of evaluative
  judgements partly based on PIs. (CHI its
  successor responsible for independent publication
  of NHS PIs from 2003.)
• The exercise of institutional (and in practice
  personal managerial) incentives and sanctions
  partly based on PI scores, eg traffic lights re
  Trusts announced 2000, subsequently star system.

4
PIs in the English NHS the story - 3

• Announcement (2004) that Healthcare Commission
  will move to balanced scorecard, distinguishing
  24 core and 10 developmental standards, the
  latter described by Secretary of State as
  aspirations not targets.
• The introduction of micro-level financial
  incentives for good PI scores via primary care
  Quality Outcomes Framework in 2004 GP contract.
• The use of PI data as a source of information to
  guide patient choice of service provider from
  2004.

5
Path-dependence theory the original perspective

• How is it possible for sub-optimal technologies
  to survive in the long run in a competitive
  market?
• Textbook examples are QWERTY keyboard (which is
  ergonomically sub-optimal) and VHS videocassette
  (which anoraks consider to be inferior to the
  Betamax system)
• Path-dependence likely to occur where
• Large set-up or fixed production costs (so
  producers want a long run)
• Individual users get more out of the technology
  with repetition (hence they have a learning
  investment in it)
• Benefits to users increases as the number of
  users increases (keyboard skills transferable, we
  can swap videos, a big market may bring more
  related products eg more videos)
• Users desire a device that is reasonably
  future-proof (Arthur, cited in Pierson 2004)

6
Core elements of PDT in public policy history

• Endogenous critical juncture, variously
• Contingent, eg wars, disasters, economic shocks
  (Hall Taylor 1996)
• Predictable policy windows eg elections
  (Kingdon 1984)
• Small events can have great impact early in a
  path (Pierson 2004) accidents can lead to major
  policy change (Klein Lewis 1976 Harrison 1994)
• Creation of new path
• Pre-existing ideas are a necessary condition
• Possible role of epistemic communities (Haas
  1989) /or policy entrepreneurs (Kingdon 1984)
• New path obscures alternative possibilities
• Costs of switching policy increase over time
• Cognitive dominance of new paths underpinning
  ideas

7
NHS PIs the critical juncture?

• Conservative govt of 1979 made few immediate
  changes to NHS, but political parliamentary
  pressure grew in 1980-81, especially related to
  growth distribution of NHS workforce
• Public Accounts Committee report of 1981 critical
  of lack of performance management by DHSS, but
  merely called for a better upward flow of
  information
• DHSS immediate response was to mention Korner
  recommendations (incorrectly implying that this
  would be performance information)
• Norman Fowler (Sec of State) Sir Kenneth Stowe
  (Permanent Secretary) succeeded Jenkin Nairne
• Stowe proposed Regional review process
• A private office DHSS civil servant invented idea
  of PIs
• DHSS took over the indicators developed by John
  Yates at Birmingham University for quite a
  different purpose

8
Positive feedback

• New path may provide power to change rules of
  the game hence self-reinforcing
• New path may lead to creation of new institutions
  with vested interests in maintaining the path
• New path may create new roles identities
• New path may lead to adaptive expectations
  modified assumptive worlds (Young 1977) so that
  it becomes taken for granted (Hall Taylor
  1996 Pierson 2004)
• New path is not static, but developments remain
  within the core ideas of the path (Steinmo 2003
  also Lindblom 1979 on incremental analysis)

9
Positive feedback NHS PIs possible examples

• Abandonment of high-profile public policy is
  potentially politically costly eg abandoning
  PIs would look like giving up or covering up
• PIs feed on themselves, eg
• An actor who is challenged in terms of PIs may
  legitimately respond in (some version) of the
  same currency, whereas a radical critique may not
  be seen as legitimate
• Tyranny of light (Tsoukas 1997) more
  information tends to create greater distrust
• Greater distrust leads to tighter coupling
  between PIs incentives/ sanctions
• Regulatory institutions (most notably CHI/
  Healthcare Commission) have vested interests in
  PIs

10
(No Transcript)
11
Some problems criticisms

• What is PDT for? (frequent concept stretching)
• In economics, PDT addresses a theoretical
  aberration, ie long-run survival of suboptimal
  technology in markets nothing like this in other
  social sciences, where suboptimality (assuming
  you can say what it is!) is probably the norm
  (Bridges 2000)
• To say that history matters (eg in opposition
  to ahistorical behavioural political science
  presumably Piersons 2000, 2004 project)
• Desire of historical institutionalists to
  explain both continuity change
• Is PDT a falsifiable theory, a heuristic for data
  collection, or just a metaphor (of paths or
  trees)?
• Problems of specification without a clear
  concept of the policy, we cannot identify a
  critical juncture, or differentiate continuity
  from change