The share of the global population residing in urban areas stood at 55 % in 2018 and is expected to increase to 68 % by 2050.1 Given such trends, assessing potential geographical patterns in the prevalence of disability within countries is crucial for informing programs, policies, and strategies towards disability-inclusive urban development.

There has been emerging evidence that disability prevalence is lower in urban areas than in rural areas. While a global study found higher disability prevalence rates in rural areas than in urban areas,2 results are more mixed in country studies. In the United States, there is a higher prevalence of disability in rural (noncore) counties than in large metropolitan centers, even after adjusting for age, sex, race/ethnicity, education, and poverty level.3 Rural residents are also more likely to report multiple disabilities and specific types such as hearing and cognitive disabilities.3,4 In China and India, older adults and women in rural areas more often have disabilities than their urban counterparts.5, 6, 7 In Bangladesh, disability prevalence is higher in rural communities at all ages.8 In Morocco, rural residence is associated with a higher prevalence of disability, particularly among older adults and those with lower education.9 Yet results from other countries such as Afghanistan suggest that prevalence is higher in urban areas compared to rural areas.10

Interpreting this literature is challenging as studies use different measures for disability and for urban development. The main goal of this paper is to estimate whether the prevalence of disability measured through functional difficulties in low- and middle-income countries (LMICs) is negatively associated with urban development, exploiting recent and internationally comparable data on functional difficulties and on satellite-generated village-level nightlight data. We contribute to the literature by producing internationally comparable estimates of the association between urban development and functional difficulties in 15 LMICs: Cambodia, Haiti, Jordan, Kenya, Mali, Mauritania, Nepal, Nigeria, Pakistan, Rwanda, Senegal, South Africa, Tanzania, Timor-Leste, and Uganda.

Our study overcomes a unique challenge. There is no universal method for the delineation of urban and rural areas. Rather, countries develop and implement their own urban-rural classification criteria. Hence, we cannot rely on each country's distinct classification of villages into urban and rural areas to assess and compare the relationship between functional difficulties and urban development across countries. To enhance the cross-country comparability of our results, we leverage village-level satellite-generated nightlight composite data, which is often used in the empirical literature as a proxy for subnational urbanization, economic activity,11, 12, 13, 14, 15 and human development.16

The relationship between disability and urban development is complex and expected to vary depending on how disability is defined and measured across contexts. We measure disability using the internationally tested Washington Group Short Set (WG-SS) of questions on functional and basic activity difficulties (hereafter, functional difficulties) in six domains (seeing, hearing, mobility, cognition, self-care, communication) and focus on the relationship between functional difficulties and urban development in resource poor settings in 15 LMICs.

Functional difficulties may be influenced by the interaction between health conditions, personal characteristics, resources, and the environment17,18.i In this context, rural-urban disparities in functional difficulties may be determined by a variety of demographic, health, socioeconomic, and environmental factors. Rural-urban disparities in functional difficulties may reflect differences in the age distribution of the population and in disease patterns across urban and rural areas. In rural areas, higher levels of poverty and poorer living conditions, such as unsafe sanitation and water, may contribute to greater prevalence and severity of health problems and functional difficulties compared to urban areas. Limited access to health resources (healthcare, rehabilitation services, and assistive technology) in rural areas is also expected to drive prevalence up. In urban areas, access to healthcare may help prevent, treat, or manage the health conditions (e.g. diabetes) that lead to functional difficulties (e.g. difficulty walking).

On the other hand, there are several reasons why functional difficulties may be more prevalent in urban areas than in rural areas. Lower survival rates for persons with functional difficulties in rural areas may lead to lower prevalence of functional difficulties. A lack of access to health information in rural areas may lead to limited awareness and/or higher stigma surrounding functional difficulties, which may reduce respondents' reporting of functional difficulties in survey data.19

Additionally, environmental factors such as infrastructure could affect mobility difficulties differently in rural and urban areas in context specific ways. For example, people with a given impairment such as arthritis may experience more difficulties walking around if, in a given country, urban areas have more physical obstacles (e.g. sidewalks without curb cuts, stairs to enter buildings) than rural areas. On the contrary, some urban areas may offer accessible infrastructure while rural areas have rugged and uneven terrain that is difficult to walk around.

Further complicating the relationship between functional difficulties and urban development, people can migrate from rural to urban areas, or vice versa. If persons with functional difficulties are less likely to migrate to urban places,20 the prevalence of functional difficulties will be higher in rural areas compared to urban areas even if the risk of functional difficulty is the same in both urban and rural areas. At any given time, the relationship between functional difficulties and urban development is a function of any underlying risk as well as a function of migration patterns.

This paper does not identify the causal links in this complex relationship between functional difficulties and urban development. Instead, it measures the correlation between functional difficulties and urban development using internationally comparable data on both functional difficulties and urban development. This correlation is important as it documents relative needs for services in different types of communities.