Research stages


The construction of the Population Health Index (INES), a consistent and comprehensive measure of a population’s health, involved a socio-technical approach that combined the MACBETH multicriteria approach (Measuring Attractiveness by a Categorical Based Evaluation Technique) and participatory methods.

The participatory methods involved Delphi panels made up of a variety of experts representing diverse points of view on the population’s health (multidisciplinary group – see Research Team and Participating Institutions) and the holding of decision conferences.

The Research Plan was carried out over a series of stages:

1. Literature review

The research began with a review of the relevant literature as a way to establish a conceptual and methodological framework that would allow for:

  • Reflecting critically on the concepts such as the population health, health inequalities, health determinants, health in all policies and population health measurement methods and instruments;
  • Identifying the indicators which characterize the health of the population, that is to say, the determining factors that affect health, as well as the indicators (proxies) that may serve to express health outcomes.

2. Population Health Index construction

The second stage, concerning  the Population Health Index (INES), followed a methodological process organized into three phases: i) selecting the indicators, ii) developing the multicriteria model for determining the health score of the Municipalities, and iii) the application of INES over a period of 20 years.
2.1. Selection of indicators

Following the initial review of the literature, the next step was to take into consideration the broad range of indicators that has come to be identified over the last few decades as i) determinants/factors influencing health (or disease) both directly and indirectly, and ii) health outcomes. From this phase of research, an initial list of 177 indicators was drawn up.

The selection of those indicators which would then comprise the INES Index was carried out by using Delphi method rules (Linstone et al, 2002; Landeta, 2006), based on a structured, iterative process of gathering and synthesizing the opinions of a group of experts. The list of indicators identified in the literature was subjected to the scrutiny of a panel, composed byGeoHealthS project experts, who considered the following selection criteria:

  • Relevance of the indicator for the assessment of population health;
  • Quality of data from the indicator, including availability and reliability of the data for a specific period of time (1991, 2001, 2011) and the study’s geographical disaggregation (municipal level).

After 3 rounds of consensus building, 62 indicators were chosen which characterize the health of the population in various dimensions, taking into consideration the:

Health determinants:

  • Economic and social factors
  • Physical Environment
  • Health Care
  • Lifestyles

Health outcomes:

  • Mortality
  • Morbidity

These dimensions correspond to the broad areas of concern for population health.

2.2. Multicriteria model

To support the construction of INES, a multicriteria model was developed with the aim of determining the value of population health for each municipality in each evaluation criterion, area of concern and in overall terms.

The multicriteria model was developed within the scope of the doctoral thesis developed by Teresa Rodrigues, researcher at IST – The Technical University of Lisbon, whose principal academic adviser was Carlos Bana e Costa (IST/UL), assisted by Mónica Oliveira (IST/UL) and Paula Santana (University of Coimbra). The MACBETH multicriteria model (Measuring Attractiveness by a Categorical Based Evaluation Technique) was used in creating this model.

MACBETH is a multicriteria approach that permits to evaluate the health of population at the municipality level on multiple criteria. This approach was chosen as it only requires qualitative judgements about differences of attractiveness to evaluate each municipality in each evaluation criterion and to weight criteria and areas of concern. MACBETH was supported by a combination of participatory methods including a modified Delphi process with a multidisciplinary group of experts, and decision conferences with a strategic group.

To build the multicriteria model, the following activities were developed:

2.2.1 Structuring

Structuring activities allowed to:

  • Define the evaluation criterion to be included in INES (evaluation criterion are grouped in several areas of concern);
  • Assign a descriptor of performance to each evaluation criterion (this was supported by the indicators selected in section 2.1).

2.2.2 Evaluating

Evaluation activities with MACBETH for:

  • Building value functions that will permit to convert the performance of a municipality in each criterion into value;
  • Determining criteria and area weights, by asking for qualitative judgements from experts.

The MACBETH multicriteria approach is supported by the M-MACBETH (www.m-macbeth.com) and WISED (www.wiseds.com) decision support systems.

More information:

Teresa C. Rodrigues, Carlos A. Bana e Costa, Mónica D. Oliveira, Paula Santana (2014)Metodologia de apoio multicritério à construção do INES (Índice do Estado de Saúde), inSantana, P. e Nossa, P. (Coord.), A Geografia da Saúde no cruzamento de saberes, Coimbra, 21-24 Abril, Grupo de Investigação em Geografia da Saúde/CEGOT

Teresa C. Rodrigues, Mónica D. Oliveira, Carlos A. Bana e Costa, Paula Santana (2014) The MACBETH approach to health value measurement: a multicriteria model for building a value-based population health index, Working Paper of Centre for Management Studies of Instituto Superior Técnico, 01/2014

Teresa. C. Rodrigues (2014) The MACBETH Approach to Health Value Measurement: Building a Population Health Index in Group Processes, Procedia Technology, volume 16, pp.1361-1366.

2.3. Applying the Population Health Index to Portuguese Municipalities

The INES makes it possible to monitor and assess the health of each municipality’s population by studying six dimensions which correspond to the main areas of concern in terms of health – socio-economic conditions, physical environment, lifestyles, healthcare services, mortality and morbidity – bringing together 43 evaluation criteria to which one or more indicators are assigned, describing the performance of each municipality with respect to each criterion.

The Population Health Index (INES) was applied to data from 278 municipalities in Portugal mainland at three moments in time (1991, 2001, and 2011).

The results obtained from the multicriteria model enable the monitoring and assessment of the health of the population of each municipality over a 20-year period of time, taking into consideration the INES score for the following outputs:

  • Performance profile
  • Score for each evaluation criterion
  • Score for each evaluation dimension and area of concern
  • Global score for the population health

These results serve as a point of departure so that political decision-makers may:

  • Analyse and make comparisons on the health of the population and inequalities amongst municipalities;
  • Identify the main problems and the priority areas of intervention, and
  • Make informed decisions in terms of promoting health within the population.

3. WebGIS platform

For the presentation and dissemination of the results, a public access WebGIS platform was created, a technological platform founded on geographic information systems.

The platform saudemunicipio makes available the maps and geographical information of INES results from Portuguese municipalities over the last 20 years. It also allows users to visualize, analyze and compare the Index outputs on the municipal level, be it the global score, the score for each evaluation dimension and area of concern, or the score for each evaluation criterion.

Link: http://saudemunicipio.uc.pt/