Population preferences and choice of primary care models A discrete choice experiment in Sweden

1
Population preferences and choice of primary care
models A discrete choice experiment in Sweden

• Jonas Hjelmgren, Anders Anell
• The Swedish Institute for Health Economics (IHE)
• P.O. Box 2127, S-220 02 Lund
• aa_at_ihe.se

2
IHE
3
Background and objective

• Several competing models of primary care in
  Sweden. Important differences in terms of choice
  options for the population.
• How important is choice of provider? How
  important is choice of GP?
• Preferences across population related to
  organisation of primary care should be of
  interest to policy-makers in several countries.

4
Methodology

• Discrete choice experiment (DCE)
• Postal survey 1 600 individuals
• Questionnaire
• Background information (age, sex, education,
  health status (EQ-5D), working situation,
  experiences with primary care)
• Assessment of preferences for attributes (DCE)
• Multi-logistic regression model was used to
  assign preference weights for the attributes

5
Selection of attributes
6
Design of experiment

• Type A
• Registration with GP
• Large influence
• Geographical responsibility
• 200 SEK user charge
• 2 days waiting time

• Type B
• Registration with PCT
• Limited influence
• Choice of provider
• 300 sek user charge
• 4 days waiting time

Choice of A, B or equal?
7
Data

• Number of respondents 924 (58)
• Mean age 48.9
• Women/Men () 41/59
• Bias towards more educated individuals (35
  university level)

8
Results importance of attributes
plt0.001 ns0.1871
9
Results preferences related to PCT
plt0.05
10
Conclusions

• Individuals trade-off attributes differently
  preferences vary systematically
• Choice of provider, influence and waiting times
  are generally important
• Choice of GP more important for elderly, the
  unhealthy and individuals pre-registered with GP
• Maintaining a system of choice and multiple
  organisational models (both GP and PCT) supported
  by expressed population preferences