A diagnostic modeling study evaluating long-term care needs across eight Middle East and North Africa countries indicates that population-level care dependency varies widely, ranging from approximately 3 percent in Oman to 22.8 percent in Saudi Arabia. Published in Population Health Metrics by Mohamed Ismail, Priyanka D. Kanth, and Shereen Hussein, the research addresses a substantial public health planning dilemma in regions where comprehensive microdata and longitudinal household surveys on functional dependency are virtually absent. Rather than relying on single indicators such as general disability prevalence or aggregate disease diagnoses, the researchers constructed a population-based Long-Term Care Needs Index designed to capture how demographic age structures, manifest disability, and major non-communicable diseases jointly generate care dependency across national populations.
The analytical framework was motivated by the intersection of rapid demographic transition and high burdens of non-communicable illnesses across the Middle East and North Africa, where the population of individuals aged 60 and older is projected to triple by 2050. Despite this swift transition, formal long-term care infrastructure remains fragmented and limited throughout the region, leaving between 80 and 95 percent of daily assistance to unpaid female family members within informal household arrangements. Because large-scale surveys measuring specific limitations in activities of daily living and instrumental activities of daily living do not exist across most Middle Eastern countries, health planners have lacked standardized methods to quantify the magnitude of potential dependency, frequently leading to under-reporting and fragmented service planning.
To overcome this data scarcity, the team selected eight countries reflecting varied income categories, demographic phases, and geographical subregions: Saudi Arabia, Morocco, Tunisia, Egypt, Lebanon, Jordan, Oman, and Qatar. The sample includes North African nations experiencing relatively advanced population ageing alongside Gulf Cooperation Council states with comparatively younger overall populations undergoing rapid epidemiological shifts. National and regional data sources, including census records, United Nations Economic and Social Commission for Western Asia reports, national health surveys, and the Global Burden of Disease 2021 estimates released in 2024, were harmonized across standardized age categories to provide unified baseline parameters for the analytical model.
Methodologically, the Long-Term Care Needs Index operates as a population-weighted linear formulation incorporating four primary elements: demographic age distribution, recorded disability prevalence, five major non-communicable diseases, and interaction terms between disability and each chronic disease. The epidemiological components focus specifically on cardiovascular disease, diabetes, cancer, Alzheimer’s disease and other dementias, and Parkinson’s disease. These conditions were selected due to their high regional prevalence and their established clinical pathways toward functional limitation. Interaction terms were introduced to mathematically approximate the compounding functional deficits that arise when individuals live with both an existing physical impairment and a chronic progressive illness, rather than treating these conditions as mutually exclusive health states.
To maintain internal mathematical validity and cross-country comparability without double-counting overlapping risk factors, the model constrains the sum of all component weights to one in a convex combination. Baseline weight parameters were derived from empirical literature and clinical evidence regarding transition-like probabilities from chronic disease and functional impairment to active care dependency within a one-year timeframe. For forward projections from 2024 through 2030, age distributions vary annually based on demographic data from the United States Census Bureau, whereas age-specific disease and disability rates are held constant at 2024 levels. This parameter design ensures that projected trajectory shifts over the six-year analytical horizon are driven specifically by demographic ageing under an expansion of morbidity assumption.
The study’s factor attribution analysis demonstrated that recorded disability serves as the predominant driver of estimated care dependency in nearly every examined country, accounting for 67 to 94 percent of total index scores in 2024. In Saudi Arabia, disability represented 93.89 percent of the total index value, while in Morocco it accounted for 88.89 percent, and exceeded 80 percent in Egypt, Tunisia, and Jordan. Even in countries with lower disability attribution, such as Lebanon at 70.33 percent and Oman at 67.08 percent, functional limitations remained the primary component. These empirical results reinforce global epidemiological observations that manifest functional restrictions, rather than chronological age alone, dictate the volume of support required for fundamental daily functioning.
A notable exception to this general pattern appeared in Qatar, where the attribution decomposition revealed a distinctly disease-driven dependency profile. In Qatar, diabetes accounted for 38.68 percent of the total care needs index, surpassing the 33.02 percent contribution from recorded disability, while cardiovascular diseases contributed an additional 19.24 percent. This contrast reflects Qatar’s unique epidemiological environment, where nearly one in six adults lives with diabetes and chronic metabolic disorders pose substantial risks of severe secondary complications such as functional decline, amputations, and cognitive impairment. The authors emphasize that this divergence illustrates why uniform regional care policies are inappropriate, requiring Gulf states to combine social support systems with aggressive metabolic disease management.
Multi-year projections tracking the index through 2030 show a consistent upward trajectory across all eight countries, generating an average annual increment of 0.002 to 0.004 index points and a cumulative increase of 10 to 15 percent over the projection period. By 2030, the projected proportion of the population requiring long-term care is estimated to reach 0.233 in Saudi Arabia, 0.195 in Morocco, 0.109 in Tunisia, 0.089 in Egypt, 0.083 in Lebanon, 0.075 in Jordan, 0.031 in Oman, and 0.015 in Qatar. In absolute terms, even modest fractional increments represent hundreds of thousands of additional individuals developing potential care needs within health systems currently characterized by limited public service capacity.
To evaluate parameter stability and the propagation of uncertainty, the investigators executed a comprehensive suite of sensitivity evaluations using data from Tunisia, comprising Bayesian inference, non-parametric bootstrapping, probabilistic perturbations, and deterministic one-at-a-time variations. In the Bayesian analysis implemented using Hamiltonian Monte Carlo in Stan, component weights were assigned Dirichlet priors, revealing that disability and its interaction terms maintained the largest posterior means. In the probabilistic perturbation assessment, injecting random noise into weights generated a tightly bounded index distribution with a median of 0.1007 and a 95 percent confidence interval spanning 0.0953 to 0.1061. Deterministic perturbations varying weights by plus or minus 50 percent confirmed that the overall model is primarily sensitive to the disability component, remaining resilient against variations in lower-weighted disease categories.
The researchers also validated the model externally by applying its mathematical framework to data from England for adults aged 55 and older. The resulting model calculation produced an estimated care dependency rate of 24.44 percent when incorporating age adjustments, aligning closely with empirical findings from the English Longitudinal Study of Ageing and the Health Survey for England, which document that approximately 24 to 28 percent of older English adults require assistance with daily living activities. While the authors emphasize that aggregate index modeling cannot substitute for future longitudinal household surveys, they note that this diagnostic framework offers a reliable, transferable planning instrument for data-scarce settings in the Middle East, Sub-Saharan Africa, and South Asia confronting accelerating population ageing.
The research also highlights pronounced discrepancies and potential under-reporting within the underlying disability data available across the region. In Tunisia, for instance, documented disability prevalence spanned a wide range from 2.15 percent to 16.3 percent, a disparity the authors attribute to social stigma, varied classification systems, and segregated education frameworks. Across the eight studied countries, recorded disability levels generally remained substantially lower than those observed in mature monitoring systems; by comparison, the 2021 Census in Wales recorded disability among 22.3 percent of women and 19.8 percent of men. While UN-ESCWA compilations estimate a disability prevalence of roughly 31 percent among older adults in Egypt, systematic national data collection capturing functional status and specific care arrangements remains restricted to Egypt and Lebanon.
Several methodological parameters and boundaries govern the model’s interpretation. The authors intentionally excluded broader socioeconomic indicators from the index formulation, reasoning that while economic factors influence healthcare access and service utilization, they do not directly trigger care dependency within the model’s one-year analytical horizon. The framework estimates underlying potential need rather than expressed service demand, which remains heavily filtered by cultural expectations, family structures, and service availability. Additionally, reliance on national aggregate figures inevitably conceals intra-country disparities, including rural-urban divides and socioeconomic gradients. Because heterogeneous national sources lacked uniform age-specific variance measures for disability, the investigators could not apply formal probabilistic uncertainty distributions to disability prevalence, though the index demonstrated stability across multiple deterministic and bootstrap sensitivity tests.
In placing the findings into global and policy context, the authors note that while universal state provision characterizes Nordic systems and hybrid insurance models operate in Japan and South Korea, Middle Eastern systems remain in an early formative stage. The stark contrast between disability-dominated profiles in North Africa and disease-driven demand in the Gulf requires differentiated national strategies. In settings like Qatar, where uncontrolled diabetes could absorb nearly one-third of national healthcare spending by 2050, the authors recommend coupling early risk screening with multidisciplinary disease management. Across the wider region, the researchers conclude that health ministries must develop longitudinal registries of functional limitations, integrate chronic disease management into social care planning, and establish formal support mechanisms for informal family caregivers.
Subject of Research: Diagnostic modeling of population-level long-term care needs in data-scarce countries across the Middle East and North Africa
Article Title: Estimating long-term care needs in data-scarce settings: a diagnostic model with evidence from MENA
Article References: Ismail, M., Kanth, P. D., & Hussein, S. (2026). Estimating long-term care needs in data-scarce settings: a diagnostic model with evidence from MENA. Population Health Metrics, 24(1), Article 42. https://doi.org/10.1186/s12963-026-00477-2
Image Credits: AI Generated
DOI: 10.1186/s12963-026-00477-2
Keywords: long-term care, care dependency, Middle East and North Africa, population ageing, disability prevalence, non-communicable diseases, cardiovascular disease, diabetes, diagnostic modeling, LTC Needs Index, Saudi Arabia, Qatar, Morocco, Tunisia, health metrics, demographic transition, informal caregiving
News Source: Ophelia Keating. (November 20, 2026). Diagnostic Model Predicts Rising Long-Term Care Needs Across Eight MENA Countries. Scienmag.



