Skip to main content
NCBI home page
Canadian Journal of Public Health = Revue Canadienne de Santé Publique logoLink to Canadian Journal of Public Health = Revue Canadienne de Santé Publique
. 2024 Jul 23;115(5):825–833. doi: 10.17269/s41997-024-00903-8

A multicriteria vulnerability index for equitable resource allocation in public health funding

Irène Abi-Zeid 1,✉, Nicole Bouchard 2, Morgane Bousquet 1, Jérôme Cerutti 3, Sophie Dupéré 4, Julie Fortier 5, Roxane Lavoie 3, Isabelle Mauger 5, Catherine Raymond 6, Estelle Richard 7, Lynda Savard 5
PMCID: PMC11559502  PMID: 39042212

Abstract

Setting

This paper describes an action research project with the Centre universitaire intégré de santé et de services sociaux - Capitale Nationale (CIUSSS-CN) who identified a need to assess vulnerability in their territories in order to ensure equitable distribution of the Integrated Perinatal and Early Childhood Services (SIPPE) program funds. The objective was to design and validate a multicriteria model to provide a more accurate portrait of vulnerability based on recent social realities.

Intervention

Our multidisciplinary research team of 7 members included experts in analytics, decision aiding, and community and public health. In collaboration with 6 CIUSSS-CN professionals, we co-constructed, during 9 workshops, a multicriteria model to aggregate the multiple dimensions of vulnerability. We used a value-focused thinking approach and applied the method MACBETH assisted by a geographic information system.

Outcomes

Criteria, scales, and weights were validated and led to a vulnerability score for each CIUSSS-CN territory. This score provides a more accurate portrait of territorial disparities based on data and the participants’ experience. The model was implemented in a dynamic user-friendly tool and serves to support decision-makers in the resource allocation process. Knowledge transfer was conducted during and after the process.

Implications

This multidisciplinary research has served to anchor public health funding in local realities, with an emphasis on equity and stakeholder engagement. Our mixed-method approach integrating qualitative and quantitative data is adaptable to other contexts. Our results can enhance intervention effectiveness and allow for a better response to the needs of the population targeted by the SIPPE program.

Supplementary Information

The online version contains supplementary material available at 10.17269/s41997-024-00903-8.

Keywords: Multicriteria evaluation, Mixed research, Vulnerability assessment, Budget allocation, Public health funding, Deprivation

Introduction

The health network in the province of Quebec is structured around Integrated (University) Health and Social Services Centres (CI(U)SSS). Their mission is to improve the overall health of the population living on their territories by ensuring accessibility and quality of health services through local community service centres (CLSC). Quebec has long invested in the Integrated Perinatal and Early Childhood Services (SIPPE) program designed to optimize child development and the health and well-being of parents with infants aged 0 to 5 (MSSS, 2019).

The CIUSSS-CN (Centre universitaire intégré de santé et de services sociaux - Capitale Nationale) in the Quebec City region caters to the needs of more than 766,579 inhabitants over an area of 18,643 km2. In recent years, it has become evident to its Public Health department that the distribution of SIPPE funds among the CLSCs must be revisited. The initial distribution, established over 15 years ago, lacked consideration for recent social realities, such as an influx of young refugee families, changes in birth rates, escalating housing costs, and concentrations of families with an Indigenous identity. To ensure equity, it had become imperative to rethink the allocation of funds based on the actual territories’ vulnerability. This led to an action research project, initiated in 2020, involving a multidisciplinary research team, including experts in analytics, decision support, community and public health, and six CIUSSS-CN participants, three of whom were from the Public Health department, while the other three were community organizers designated as spokespersons for community organizations in the field. The project was necessary to fill an important gap since none of the existing deprivation indices adequately captured the vulnerability of the SIPPE population. The existing indices, such as the Deprivation Index (Pampalon et al., 2009) or the Canadian Index of Multiple Deprivation (Matheson et al., 2012) are, according to community organizers on the field, too general and too detached from the territories’ reality. Furthermore, they are not detailed enough to support funding decisions.

The objective of this project was to design and apply a model for evaluating the SIPPE population’s vulnerability to ensure an equitable allocation of funds between the CLSC territories. The role of the CIUSSS-CN participants was to provide the knowledge needed to structure the problem, build the vulnerability evaluation model, and validate it. During the project, they continually relayed information on the project’s progress to other actors and acted as spokespersons for their concerns and needs. The Université Laval research team acted as an impartial facilitator. The overall socio-technical intervention to build, validate, and implement a multicriteria vulnerability evaluation model required nine online workshops of 3 h over a period of 8 months between September 2021 and April 2022.

Methods

Our research process consisted of six main steps adapted from Abi-Zeid et al. (2023): (1) structuring the problem; (2) constructing the criteria and their scales; (3) weighting the criteria; (4) computing the territories’ global vulnerability scores; (5) validating the model and implementing the results in a decision support tool; and (6) making recommendations regarding funds allocation. Knowledge transfer was continuous during the whole process. We used the multicriteria MACBETH method (Bana e Costa et al., 2012) to obtain, using a weighted sum, the global vulnerability scores.

Multicriteria decision aiding

Multicriteria decision aiding consists of families of methods aiming at constructing and explicitly considering multiple heterogeneous, quantitative, or qualitative criteria to help individuals or groups in their evaluation and decision process (Belton & Stewart, 2002). Socio-technical MCDA processes, in which an MCDA model is co-constructed, include considerations related to the organizational context and environment to foster dialogue and transparency, thus increasing acceptance of the results (Phillips & Bana e Costa, 2007). MCDA is a well-established field with numerous applications, including healthcare (Khan et al., 2022).

We selected the MACBETH method for several reasons. First, it avoids the main traps associated with aggregating several criteria to obtain a score based on a weighted sum (Bana e Costa et al., 2012). Despite being regularly used in practice, the weighted sum often leads to methodological problems such as using non-cardinal scales or misinterpreting the meaning of weights. In fact, this is the most common and critical mistake in weighted sums. It is important to note that criteria in a weighted sum do not reflect the importance of criteria; they depend on the measurement ranges (Keeney, 2002). Second, MACBETH is relatively simple for participants since it uses semantic scales to construct cardinal value functions for the criteria. Third, its software (M-MACBETH) allows simple and effective recording of the judgements provided by participants and can detect inconsistencies.

To construct a cardinal value scale for a criterion using the MACBETH method, participants define references chosen among the criterion’s possible values, where, by convention, lower references have a value of 0 (lower vulnerability), and upper references have a value of 100 (higher vulnerability). As an example, the immigration criterion in our model is measured as the percentage of children 0–5 years old living in a family where at least one parent is a recent non-permanent resident. The participants defined the performances corresponding to a lower reference and an upper reference as 10% and 15%, respectively. This means that a territory with a performance of 10% could be considered slightly vulnerable from the perspective of this criterion. In contrast, a performance of 15% makes a territory highly vulnerable. The scales obtained are open allowing for the possibility of negative vulnerability values or values exceeding 100.

Once the performances on a criterion are ranked in decreasing order of vulnerability (from 30% to 0% in Figure 1), the next step is to ask the participants to qualify the difference in vulnerability between pairs of performances using a semantic scale (extreme, very strong, strong, moderate, weak, very weak, null). For example, Figure 1 illustrates that the difference in vulnerability between a territory with 5% and one with 10% of children from immigrant families is considered, by the participants, to be moderate. In comparison, the difference between 25% and 30% is considered high.

Fig. 1.

Fig. 1

Open in a new tab

Judgement matrix in the M-MACBETH software for the immigration criterion

Subsequently, the M-MACBETH software provides a cardinal scale that can be adjusted manually if it does not contradict the judgement matrix. We see, for example in Figure 2, that a territory with 20% of the children in immigrant families has a vulnerability score on this criterion of 250 and a territory with 5% of the children in immigrant families has a vulnerability score of −150.

Fig. 2.

Fig. 2

Open in a new tab

Cardinal value scale for the Immigration criterion

To obtain the scaling constants (weights) for the weighted sum, a similar procedure is followed. Initially, a set of fictitious territories is created in which the performances are lower references for all criteria except for one criterion, for which the performance is the upper reference. The fictitious territories are then compared pairwise using the same semantic scale above. Subsequently, the software computes the scaling constants which can also be adjusted as long as they do not contradict the judgement matrix.

Results

During the first two workshops (problem structuring/methodology step 1), the discussions evolved around the following questions: “If there were a truly vulnerable territory, what would its characteristics be? What would make it vulnerable? Why? What would the characteristics of a non-vulnerable territory be? Why?” Vulnerability, in this context, pertained to risks that could impact the development and well-being of children and parents from SIPPE populations. A comprehensive list of over 80 elements emerged from these discussions. Based on these elements, we formulated 10 criteria detailed below.

Economic status dimension

Criterion C1, “Low Income”, corresponds to the percentage of children aged 0–5 years old living in a low-income family defined according to the market basket measure from Statistics Canada. It ranges between 3% and 22% with the lower reference at 7% and the upper reference at 10%.

Access to services dimension

Criterion C2, “Access to services”, assesses the accessibility to basic services such as schools, health care, public transportation, food, recreational activities, and social organizations. It is used as a proxy for families’ isolation levels and their reliance on cars, which can strain family finances. To design this qualitative criterion, the participants used (1) their experience with the territories, (2) the walk score (Hall & Ram, 2018), and (3) the Canadian Urban Environmental Health Research Consortium maps. A qualitative performance scale from 1 to 6 was constructed, and each territory was assigned to one of the levels. The lower reference was set at 3 and the upper reference at 4.

Socio-demographics dimension

Criterion C3, “Immigration”, looks at the percentage of children aged 0–5 years old with at least one parent classified as a non-permanent resident between 2011 and 2016. This includes international students, temporary workers, and refugees. It ranged from 0% to 30% with the lower reference at 10% and the upper reference at 15%.

Criterion C4, “Indigenous identity”, looks at the percentage of children 0–5 years old with at least one parent with an Indigenous identity. It ranged from 0% to 12% with the lower reference at 3% and the upper reference at 5%.

Family characteristics dimension

Criterion C5, “Single-parent families”, looks at the percentage of children 0–5 years old living with a single parent. It ranged from 0% to 50% with the lower reference at 10% and the upper reference at 15%.

Criterion C6, “Mother with only a high school diploma”, measures the percentage of births to mothers with a high school diploma at the child’s birth. It ranged from 0% to 50% with the lower reference at 10% and the upper reference at 16%.

Criterion C7, “Mother without a high school diploma”, measures the percentage of births to mothers without a high school diploma at the child’s birth. It ranged from 0% to 10% with the lower reference at 2% and the upper reference at 5%.

Criterion C8, “Mother’s age”, examines the percentage of live births to mothers aged 15–24. It ranged from 0% to 20% with the lower reference at 10% and the upper reference at 15%.

Criterion C9, “Housing issues”, looks at the percentage of children 0–5 years old living in families with housing problems (e.g. need for major renovations, insufficient size, at least 30% of family’s income goes to rent). It ranged from 0% to 40% with the lower reference at 10% and the upper reference at 20%.

Child development dimension

Criterion C10, “Child Development”, looks at the percentage of children 0–5 years old who are vulnerable in at least one dimension of development as defined in the EQDEM (Enquête québécoise sur le développement des enfants à la maternelle/Québec survey of child development in kindergarten) survey: (1) Physical health and well-being; (2) Social skills; (3) Emotional maturity; (4) Cognitive and language development; (5) Communication skills and general knowledge. A child is considered “vulnerable” when the score in a developmental domain is in the bottom 10th percentile of all scores. It ranged from 0% to 60% with the lower reference at 10% and the upper reference at 25%.

Performances of the territories on criteria C1, C3, C4, C5, and C9 were obtained using the most recent available census data from Statistics Canada, namely 2016. Values of the criteria C6 to C8 were provided by data that the CIUSSS-CN collects on its territories. C10 was extracted from the EQDEM, conducted by the Institut de la Statistique du Québec (2017). C2 is a qualitative criterion constructed by CIUSSS-CN members based on their knowledge of the territories. For all criteria, the higher the value, the more vulnerable the territory and the greater the need for funding.

The vulnerability scores

After having constructed the cardinal value scales and the weights (Supplementary Material, Figures 1 to 10), global vulnerability scores were computed for each territory (Figure 3). With an overall score of 244, the most vulnerable territory is Vanier, a neighbourhood widely recognized by community workers as deprived. In contrast, the Laurentian territory (score = −9) includes many of the wealthiest neighbourhoods in Quebec City. Charlevoix-Est (score = 181) has the highest vulnerability outside Quebec City.

Fig. 3.

Fig. 3

Open in a new tab

Global vulnerability scores

The CIUSSS-CN team validated the model, affirming its accurate representation of local realities based on their experience and field knowledge. As for the budget distribution, the recommendation was that each of the 14 territories receive a baseline amount X, to which a percentage of the remaining amount Y would be added in proportion to its vulnerability score. For example, Laurentian would receive X since it has the lowest vulnerability score, whereas Vanier would receive X + 20% × Y since 20% is the vulnerability score of Vanier divided by the sum of all the territories’ scores. To determine the value of X, meetings took place with community organizers of all CLSCs, including those who did not participate in the project, as well as with representatives of community organizations that implement SIPPE projects. After having analyzed four scenarios, they agreed on an X value and the resulting budget allocation.

To facilitate future updates of vulnerability scores and budget allocations, namely with the recent availability of 2020 Census data, we implemented the multicriteria evaluation model in an easy-to-use spreadsheet delivered to the CIUSSS-CN team and available on request from the first author.

Discussion

In the absence of an adequate index to measure, in a comprehensive manner, the vulnerability of SIPPE populations, we developed a new multicriteria vulnerability evaluation model. Some of our criteria are similar to those found in the literature in the context of deprivation indices. For the Low-income criterion, a parallel can be drawn with many authors who use average household income (Ivaldi et al., 2020; Lòpez-De Fede et al., 2016; McHenry & Rinner, 2016; Wang et al., 2021). For the Access to services criterion, we find similar indicators such as access to health services (Cabrera-Barona et al., 2016; Wang et al., 2021), access to food services (Wang et al., 2021), or access to education (Peralta et al., 2019). For the Immigration criterion, other authors have indicators that also address immigration (Cebrecos et al., 2018; Ivaldi et al., 2020). As for the Indigenous identity criterion, we did not find deprivation indicators in North America that considered Indigenous identity. However, a parallel could be made with the ethnic identity issues found in some studies in Central and South America (Ivaldi et al., 2020; Peralta et al., 2019).

Many authors include the single-parent families criterion in their indicators (Cebrecos et al., 2018; Ivaldi et al., 2020; Pampalon et al., 2012; Yun et al., 2016). For the education level criterion, our study focused on mothers. The issue of education for the general population is found in many deprivation indices (Cabrera-Barona et al., 2015, 2016; Cebrecos et al., 2018; Ivaldi et al., 2020; Lòpez-De Fede et al., 2016; McHenry & Rinner, 2016; Pampalon et al., 2009; Peralta et al., 2019; Silva & Padeiro, 2020; Wang et al., 2021; Yun et al., 2016). However, only two authors address issues related to the mother’s age at childbirth (McHenry & Rinner, 2016; Yun et al., 2016). The criterion related to Housing issues is found very often in the literature, namely in terms of healthy housing, cost, size relative to the family living in it, or the presence of essential amenities, for example toilets or kitchens (Cabrera-Barona et al., 2015, 2016; Ivaldi et al., 2020; Lòpez-De Fede et al., 2016; McHenry & Rinner, 2016; Peralta et al., 2019; Silva & Padeiro, 2020; Wang et al., 2021; Yun et al., 2016). The Child development criterion is unique to our study.

In order to compare our vulnerability index with existing indices in Quebec such as the Deprivation Index (INSPQ, 2016) and the Canadian marginalization index (CMI) (van Ingen & Matheson, 2022), we computed correlations of our vulnerability score with the total percentage of individuals in the 4th and 5th quintile for each of the two INSPQ dimensions1 and the four CMI dimensions2. We note a strong correlation between our index and the material dimensions of both indices (77% and 92% respectively). All other correlations are moderate. The correlation with the social dimension of INSPQ’s index (0.41) is not surprising since INSPQ’s definition of social deprivation is partly based on the proportion of people aged 15 and over living alone in their home; the proportion of people aged 15 and over separated, divorced, or widowed; and the proportion of single-parent families. The first characteristic does not capture at all the vulnerability of the SIPPE population, the second characteristic only partly, while the third characteristic is actually included in our index.

The households and dwellings dimension of the CMI is computed based on the proportion of the population living alone; the proportion of the population who are not youth (age 5–15); the average number of persons per dwelling; the proportion of dwellings that are apartments in a building with 5 or more stories; the proportion of the population who are single/divorced/widowed; the proportion of dwellings that are not owned; and the proportion of the population who moved during the past 5 years. Again, the first and second characteristics do not reflect our population and the others do not correspond to the housing issues that contribute to the vulnerability of our population. This explains the moderate correlation of 0.44.

Furthermore, our index correlated moderately (0.5) with the Age and labour force dimension of the CMI that is based on the proportion of the population who are aged 65+; on the dependency ratio (total population 0‒14 and 65+/ total population 15 to 64); and on the proportion of the population not participating in labour force (aged 15+). Again, this does not reflect our SIPPE population. As for the Immigration and visible minority dimension, based on the proportion of the population who are recent immigrants (past 5 years) and the proportion of the population who self-identify as a visible minority, it correlates the least (0.32) with our index that is measured from the points of view of the children living in a family with at least one recent immigrant.

Our co-constructive multicriteria approach, which integrates both quantitative and qualitative data along with the value judgements of stakeholders, is innovative and a first in the vulnerability evaluation literature. It is precise enough to be used as a basis for funds allocation. By being a participative transparent approach, it distinguishes itself from other indices that predominantly rely on black-box statistical approaches and purely quantitative data such as in Pampalon et al. (2009), Land and Michalos (2018), and Matheson et al. (2012). Curiously enough, people tend to think that quantitative criteria are more “objective” than qualitative criteria. This is a trap since, in both cases, measurements need to be interpreted as a function of values. Another common misconception is that a quantitative measurement automatically defines a criterion, which is also misleading. For example, it was not easy to agree on the type of measurement to use between absolute numbers (frequencies) or relative numbers (percentages). It took two workshops totaling 6 h to reach an agreement (use percentages for criteria). It was not a matter of disagreement within the group; rather, participants found it challenging to decide individually.

The multicriteria-based approach presented is transferable to other contexts where context-based vulnerability or deprivation indices are needed. Nonetheless, some adaptation will be necessary to account for specific regional variations or unique characteristics. We strongly recommend moving away from a purely statistical approach towards multicriteria socio-technical interventions where experience and knowledge are integrated in a rigorous process based on both quantitative and qualitative criteria. Numbers must be interpreted from the point of view of those who will apply them to inform decision-making. We believe such a vulnerability index cannot be constructed in isolation. Stakeholders’ participation is crucial to ensure acceptance. Our work can serve as a reference and a starting point.

Several months after the completion of the project, the research team received the following positive feedback from the team leader of the SIPPE program: “The approach was well received by community organizers, and the strategy of informing them as we went along helped win over members for whom there was a financial loss. We had two meetings to explain the approach, and the members clearly understood the relevance of reviewing the distribution of the sum in the light of current reality. In short, the support of your team and the proposed approach were the keys to this success!”

Conclusion

The model and tool (spreadsheet) developed provide a measure of the vulnerability of the SIPPE population and enables the equitable funding of different territories, based on the values and experience of the CIUSSS-CN actors. This is a key innovation since it offers a comprehensive mechanism for distributing available funds. The spreadsheet allows for regular reviews and updates, which might be necessary to ensure ongoing relevance and accuracy in reflecting the dynamic nature of the populations and their needs. In addition to facilitating diagnosis and decision-making, our solution can also support monitoring. This is a significant contribution to the healthcare policy field. It highlights the importance of mixing data and field experience to better reflect the reality faced by service providers and beneficiaries and to anchor public health funding in local realities, with an emphasis on equity and stakeholder engagement. Our approach integrating qualitative and quantitative data is adaptable to other contexts.

Implications for policy and practice

What are the innovations in this policy or program?

  • A transparent rigorous multicriteria approach, combining quantitative and qualitative data, to compute vulnerability scores based on participants’ judgements and knowledge of the terrain. It fills a gap in the current deprivation indices, not tailored to explicitly support funding allocations.

  • A more accurate current and future portrait of vulnerability based on new social realities and disparities not reflected in current deprivation indices, and evidence-based budget allocation, which can lead to higher social acceptance.

  • Dynamic and easy-to-use tool and formula to ensure equitable distribution of funds allowing for public health interventions where they are the most needed.

What are the burning research questions for this innovation?

  • How to anchor public health funding decisions in local realities to ensure better use and equitable distribution of available public health funds?

  • How to take into account quantitative and qualitative data as well as tacit knowledge of the vulnerability that actors on the ground have?

  • How to evaluate the vulnerability of the population of interest more accurately than the current black-box statistical deprivation indices and ensure transparency of decisions?

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 154 KB) (153.7KB, pdf)

Acknowledgements

The authors wish to thank the CIUSSS-CN team for participating in the workshops and for their collaboration throughout the process. We also thank Sophie Métivier and Oscar Nilo for their technical support, and Statistics Canada personnel for their valuable help in acquiring the data.

Author contributions

No artificial intelligence (AI)–assisted technologies were used in this text nor in the project.

Abi-Zeid: conceptualization, methodology, formal analysis, original draft, writing—review and editing, supervision, project administration, funding acquisition, data curation. Bouchard: conceptualization, formal analysis. Cerutti: data curation, writing—original draft. Fortier: conceptualization, methodology, formal analysis, project administration, funding acquisition, data curation. Bousquet: data curation, writing—review and editing. Dupéré: conceptualization, review and editing, funding acquisition. Lavoie: conceptualization, methodology, formal analysis, funding acquisition. Mauger: conceptualization, formal analysis. Raymond: conceptualization, formal analysis. Richard: conceptualization, formal analysis. Savard: conceptualization, formal analysis, supervision.

Funding

This project was funded by a Partnership Engage Grant number 892-2020-3001 from the Social Sciences and Humanities Research Council.

Availability of data and material

The data must be purchased from Statistics Canada.

Code availability

The MACBETH model can be made available; however, the data must be purchased from Statistics Canada.

Declarations

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The authors declare no competing interests.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Abi-Zeid, I., Marleau Donais, F., & Cerutti, J. (2023). Multicriteria decision aiding: Challenges in real-life interventions. In: Norese, M.-F., De Vicente y Oliva, M. A., Abi-Zeid, I. (eds) Multicriteria decision aiding interventions - Applications for analysts, (pp. 161‒186). Springer Cham.
  2. Bana e Costa, C. A., De Corte, M., & Vansnick, J. C. (2012). MACBETH. International Journal of Information Technology & Decision Making,11, 2. [Google Scholar]
  3. Belton, V., & Stewart, T. J. (2002). Multiple criteria decision analysis: An integrated approach. Springer: New York, NY. 10.1007/978-1-4615-1495-4 [Google Scholar]
  4. Cabrera-Barona, P., Murphy, T., Kienberger, S., & Blaschke, T. (2015). A multi-criteria spatial deprivation index to support health inequality analyses. International Journal of Health Geographics,14(1), 11. 10.1186/s12942-015-0004-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cabrera-Barona, P., Wei, C., & Hagenlocher, M. (2016). Multiscale evaluation of an urban deprivation index: Implications for quality of life and healthcare accessibility planning. Applied Geography,70, 1–10. 10.1016/j.apgeog.2016.02.009 [Google Scholar]
  6. Cebrecos, A., Domínguez-Berjón, M. F., Duque, I., Franco, M., & Escobar, F. (2018). Geographic and statistic stability of deprivation aggregated measures at different spatial units in health research. Applied Geography,95, 9–18. 10.1016/j.apgeog.2018.04.001 [Google Scholar]
  7. Hall, C. M., & Ram, Y. (2018). Walk score® and its potential contribution to the study of active transport and walkability: A critical and systematic review. Transportation Research Part D: Transport and Environment,61, 310–324. 10.1016/j.trd.2017.12.018 [Google Scholar]
  8. INSPQ. (2016). Indice de défavorisation matérielle et sociale compilé par le Bureau d’information et d’études en santé des populations (BIESP) à partir des données des recensements canadiens de 1991, 1996, 2001, 2006, 2011 et 2016. [Online]. Institut national de santé publique du Québec. https://www.inspq.qc.ca/defavorisation/indice-de-defavorisation-materielle-et-sociale
  9. Ivaldi, E., Parra Saiani, P., Primosich, J. J., & Bruzzi, C. (2020). Health and deprivation: A new approach applied to 32 Argentinian urban areas. Social Indicators Research,151(1), 155–179. 10.1007/s11205-020-02369-w [Google Scholar]
  10. Keeney, R. L. (2002). Common mistakes in making value trade-offs. Operations Research, 50(6), 935–945. http://www.jstor.org/stable/3088614
  11. Khan, I., Pintelon, L., & Martin, H. (2022). The application of multicriteria decision analysis methods in health care: A literature review. Medical Decision Making,42(2), 262–274. 10.1177/0272989X211019040 [DOI] [PubMed] [Google Scholar]
  12. Land, K. C., & Michalos, A. C. (2018). Fifty years after the social indicators movement: Has the promise been fulfilled? An assessment and an agenda for the future. Social Indicators Research,135, 835–868. 10.1007/s11205-017-1571-y
  13. Lòpez-De Fede, A., Stewart, J. E., Hardin, J. W., & Mayfield-Smith, K. (2016). Comparison of small-area deprivation measures as predictors of chronic disease burden in a low-income population. International Journal for Equity in Health,15(1), 89. 10.1186/s12939-016-0378-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. McHenry, M., & Rinner, C. (2016). The impact of multi-criteria decision analysis parameters on an urban deprivation index. Cartographica the International Journal for Geographic Information and Geovisualization,51(4), 179–197. 10.3138/cart.51.4.3210 [Google Scholar]
  15. Matheson, F. I., Dunn, J. R., Smith, K. L., Moineddin, R., & Glazier, R. H. (2012). Development of the Canadian Marginalization Index: A new tool for the study of inequality. Canadian Journal of Public Health,103(8 Suppl 2), S12–S16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. MSSS. (2019). Les services intégrés en périnatalité et pour la petite enfance à l’intention des familles vivant en contexte de vulnérabilité : Cadre de référence, Québec, Ministère de la Santé et des Services Sociaux, p. 17. https://publications.msss.gouv.qc.ca/msss/fichiers/2018/18-836-02W.pdf. Accessed 19 Nov 2020.
  17. Pampalon, R., Hamel, D., Gamache, P., & Raymond, G. (2009). A deprivation index for health planning in Canada. Chronic Diseases in Canada,29(4), 178–191. [PubMed] [Google Scholar]
  18. Pampalon, R., Hamel, D., Gamache, P., Philibert, M. D., Raymond, G., & Simpson, A. (2012). Un indice régional de défavorisation matérielle et sociale pour la santé publique au Québec et au Canada. Canadian Journal of Public Health,103, S17–S22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Peralta, A., Espinel-Flores, V., Gotsens, M., Pérez, G., Benach, J., & Marí-Dell’Olmo, M. (2019). Developing a deprivation index to study geographical health inequalities in Ecuador. Revista de Saúde Pública, 53, 97. 10.11606/s1518-8787.2019053001410 [DOI] [PMC free article] [PubMed]
  20. Phillips, L. D., & Bana e Costa, C. A. (2007). Transparent prioritisation, budgeting and resource allocation with multi-criteria decision analysis and decision conferencing. Annals of Operations Research,154(1), 51–68. 10.1007/s10479-007-0183-3 [Google Scholar]
  21. Statistics Canada. (2021). The Canadian Index of Multiple Deprivation—User guide. Stat Canada Cat No 45-20-0001. Available at https://www150.statcan.gc.ca/n1/pub/45-20-0001/452000012023002-eng.htm. Accessed 3 Mar 2024.
  22. Silva, K. S. D. N., & Padeiro, M. (2020). Assessing inequalities in geographical access to emergency medical services in metropolitan Lisbon: A cross-sectional and ecological study. BMJ Open,10(11), e033777. 10.1136/bmjopen-2019-033777 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. van Ingen, T., & Matheson, F. I. (2022). The 2011 and 2016 iterations of the Ontario Marginalization Index: Updates, consistency and a cross-sectional study of health outcome associations. Canadian Journal of Public Health,113(2), 260–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Wang, A., Kho, A. N., Black, B., & French, D. D. (2021). Determining the feasibility of an index of the social determinants of health using data from public sources. Informatics for Health and Social Care,46(2), 205–217. 10.1080/17538157.2021.1880413 [DOI] [PubMed] [Google Scholar]
  25. Yun, J.-W., Kim, Y.-J., & Son, M. (2016). Regional Deprivation Index and socioeconomic inequalities related to infant deaths in Korea. Journal of Korean Medical Science,31(4), 568. 10.3346/jkms.2016.31.4.568 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (PDF 154 KB) (153.7KB, pdf)

Data Availability Statement

The data must be purchased from Statistics Canada.

The MACBETH model can be made available; however, the data must be purchased from Statistics Canada.

Articles from Canadian Journal of Public Health = Revue Canadienne de Santé Publique are provided here courtesy of Springer

RESOURCES