Health in Context

A Course in Multilevel Modelling for Public Health and Health Services Research

Professor Alastair H Leyland, MRC/CSO Social and Public Health Sciences Unit, Glasgow
Professor Peter P Groenewegen, Netherlands Institute for Health Services Research, Utrecht

1 July 2013 - 5 July 2013 9am – 5pm

Australian Technology Park, Meeting Room 5C

The Outcomes, Services and Policy for the Reproductive and Early Years (OSPREY) capacity building program is pleased to announce it is supporting a five day Multilevel Modelling workshop for researchers.

Multilevel modelling (MLM) is increasingly being used in public health and health services research. Pioneering development was in education, where researchers have been interested in studies examining how pupil outcomes (such as examination scores) are related to both the characteristics of the pupils themselves and those of the schools. MLM has since been widespread in the fields of health services research, epidemiology and public health, assisted by the development of specialist multilevel software and the addition of multilevel capabilities to common statistical packages.
The course includes introductory lectures, computer classes, individual consultation and two assignments. It has been delivered in 10 different countries to date, and has proved extremely popular with public health and health services researchers. It will clarify what MLM is about and why it is important to take both shared contexts and individuals into account when undertaking public health and health services research. It is important for researchers in these fields to have at least a basic understanding of this technique so as to be able to judge research that uses it, and of course also to be able to apply it themselves.

Course Teachers

Professor Alastair Leyland

Professor Leyland is an applied statistician who has been working in academic public health research for 25 years. In this time he has pioneered the use of multilevel models in health research. Professor Leyland is a Programme Leader at the MRC/CSO Social and Public Health Sciences Unit in Glasgow, Scotland, where he heads a programme of research “Measuring health, variations in health and determinants of health”, which is core funded by the Chief Scientist Office of the Scottish Government Health Directorates. He is also Honorary Professor in the Division of Community Based Sciences at the University of Glasgow.

He has developed considerable expertise in the analysis of routine and survey data, particularly using the system of Scottish Morbidity Records linking hospital discharges, cancer registrations, and death records.

Further details can be found through SPHSU (opens in a new window).

Professor Peter Groenewegen

Professor Groenewegen is a social scientist who has worked in health services research for more than 30 years. He is the Director of the Netherlands Institute for Health Services Research (NIVEL), and holds a special chair in the social and geographical aspects of health and health care in the Departments of Sociology and Human Geography at Utrecht University.

Professor Groenewegen’s team at NIVEL conducts research at the national and international level on the relationships between the need for health care, the supply of health care and health care policy. His own specific research interests include general practice, international comparative studies, health policy, health care organisation, manpower planning and the sociology of the professions. He has worked on many innovative studies using multilevel modelling, with a particular focus on exploring and understanding family influences on health-seeking behaviour.

Further details can be found through NIVEL (opens in a new window).

Course Objectives

  • to gain an understanding of the relevance of multilevel modelling for public health and health services research
  • to gain a basic insight into the theoretical backgrounds of relationships between (macro-level) contexts and (individual-level) behaviour
  • to gain an understanding of the methodological and basic statistical backgrounds of multilevel modelling
  • to learn to work with the multilevel modelling software MLwiN.

Preliminary requirements

The course is aimed at researchers who understand and are used to applying basic statistical analyses such as multiple linear regression, logistic regression and/or analysis of variance.

The computer classes will use MLwiN, a specialist multilevel modelling package. No previous knowledge of MLwiN will be assumed.

Participants will need to bring their own laptop. The minimum requirements for MLwiN are 32 Mb RAM/ 100Mb hard disk. Access more detailed system requirements (opens in a new window).

Please contact Kate Churruca k.churruca@uws.edu.au should a laptop be required for the course.

Workshop Program

  • Theoretical introduction to MLM: macro-micro relationships in health services research
  • Methodological introduction to MLM
  • Graphs, equations and multilevel modelling
  • Apportioning variation in multilevel models
  • Context and composition in multilevel models
  • Ecometrics: using MLM to construct ecological measures from individual data
  • Assignment work
  • Feedback on assignment work: critical reading
  • Computing classes

The full workshop program will be available to participants on registration.

Cost

$2,500, including morning tea, lunch and afternoon tea.

Registration

Numbers are strictly limited so placements will be allocated to the first twenty-five registrants.

To register your interest please contact Kate Churruca at the University of Western Sydney via email k.churruca@uws.edu.au or phone 02 4620 3955.

^ Back To Top