Silas Amo-Agyei

Universal Credit, one of the most significant welfare reforms in the UK in recent decades, was de- signed to simplify the benefit system and strengthen work incentives by consolidating six means- tested programmes into a single payment. We evaluate its short-term economic impacts using two complementary strategies. First, exploiting the phased geographical introduction of Univer- sal Credit across Local Authority Districts in England between 2013 and 2018, we implement a difference-in-differences design using district-level administrative data on economic activity and employment rates, weighting districts by their baseline working-age population. Second, we use individual-level panel data from the UK Household Longitudinal Study to estimate the differen- tial effect of unemployment under Universal Credit relative to the legacy system on employment, hours worked and income. At the district level, the estimated effects are small and not statis- tically significant, at −0.13 percentage points for economic activity and 0.21 for employment. At the individual level, previously unemployed individuals in areas that had introduced Uni- versal Credit experience a statistically significant reduction in monthly income of approximately 7.7%. The penalty reflects lower earnings that the benefit system does not offset, and it is not accompanied by improvements in employment or hours worked. Set alongside existing evidence on the adverse mental-health consequences of the reform, Universal Credit delivered no measur- able short-term economic gains while coinciding with declines in income security and population mental health.

We provide new evidence on the mental health and healthcare utilisation impacts of the roll-out of Universal Credit, one of the most significant welfare reforms in Europe in recent decades. Designed to simplify the benefit system and strengthen work incentives by merging six means-tested programmes into a single payment, Universal Credit also introduced features that may adversely affect mental health. Exploiting the plausibly exogenous, staggered geographical introduction of Universal Credit across England between 2013 and 2018, we implement a difference-in-differences design using comprehensive small-area administrative data on clinically recorded mental health and related healthcare utilisation. We find adverse effects of Universal Credit exposure on population mental health: a 0.10 standard-deviation increase in clinically diagnosed depression prevalence and a 0.06 standard-deviation rise in antidepressant prescribing, alongside a smaller 0.03 standard-deviation increase in mental health-related hospital admissions and attendances. In natural units, these correspond to approximately 113,742 additional cases of diagnosed depression, 29,993 extra hospital admissions and attendances, and 1.29 million additional antidepressant prescriptions annually by 2018. Results are robust across specifications, sensitivity analyses, spatial aggregation, and alternative estimators. Because outcomes are measured at the small-area level, our estimates capture population-wide effects that combine direct impacts on Universal Credit recipients with spillovers to non-recipients. Indicative valuations imply sizeable associated quality-of-life losses and direct healthcare costs of approximately £2.84 billion per year. These findings show that welfare reforms can generate substantial mental health externalities, and that the health and healthcare-system consequences of social policy belong squarely within its design and evaluation.

We provide new evidence on the mental health and healthcare utilisation impacts of the roll-out of Universal Credit, one of the most significant welfare reforms in Europe in recent decades. Designed to simplify the benefit system and strengthen work incentives by merging six means-tested programmes into a single payment, Universal Credit also introduced features that may adversely affect mental health. Exploiting the plausibly exogenous, staggered geographical introduction of Universal Credit across England between 2013 and 2018, we implement a staggered difference-in-differences design using comprehensive small-area administrative data on clinically recorded mental health and related healthcare utilisation. We find substantial adverse effects of Universal Credit exposure: a 0.10 standard-deviation increase in clinically diagnosed depression prevalence, a 0.03 standard-deviation increase in mental health-related hospital admissions and attendances, and a 0.06 standard-deviation rise in antidepressant prescribing. In natural units, these correspond to approximately 113,742 additional cases of diagnosed depression, 29,993 extra hospital admissions and attendances, and 1.29 million additional antidepressant prescriptions annually by 2018. Results are robust across specifications, sensitivity analyses, spatial aggregation, and alternative estimators. Because outcomes are measured at the small-area level, our estimates capture population-wide effects that combine direct impacts on Universal Credit recipients with spillovers to non-recipients. Indicative valuations imply sizeable associated quality-of-life losses and direct healthcare costs of approximately £2.84 billion per year. These findings highlight that welfare reforms can generate substantial mental health externalities and underscore the importance of incorporating health and healthcare system consequences into social policy design and evaluation.

The 2013 World Health Organization (WHO) guidelines for cervical pre-cancer screening recommended screening women aged 30–49 and women living with HI

Introduction: Multidimensional poverty (deprivations in education, health, and living standards) affects people with and without HIV. We compared poverty levels by HIV status in Eastern and Southern Africa and identified indicators driving deprivations. Method: We analysed the 2020–22 Population HIV Impact Assessment data from Eswatini, Lesotho, Malawi, Mozambique, Tanzania, Uganda, and Zimbabwe using the Alkire-Foster method, calculating the multidimensional poverty index (MPI), headcount ratio, and poverty intensity, using the Stata 14.2 mpi command. We classified individuals deprived in 20·0%–33·3% of indicators as vulnerable to poverty, ≥33·3% as poor, and ≥50·0% as severely poor. We estimated the number of people in each poverty category, and decomposed poverty by indicators. Analyses were survey-weighted, disaggregated by sex, residence, and age (15–24 years), and differences by HIV status tested using the Rao–Scott chi-square test (p <0·05). Results: People living with HIV (PLHIV) comprised 7·1% (11·8 million) of the study population (164·9 million). PLHIV were more likely to be vulnerable to poverty, poor, or severely poor than people without HIV. In Eswatini, with the lowest poverty level, PLHIV had higher MPIs (0·248 95% CI [0·239–0·257]) than people without HIV (0·220 [0·215–0·226]). 53·7% [51·8%–55·5%], 99,000, PLHIV were poor compared to 47·5% [46·4%– 48·6%], 266,000, of people without HIV. In Mozambique, with the highest poverty level, the MPIs were similar for people living with and without HIV, but poverty remained higher among PLHIV (70·2% [67·8%– 72·5%], 1·5 million, versus 69·6% [68·6%–70·5%], 10·5 million). The intensity of poverty did not differ across the countries. Education/employment and living standards accounted most for deprivations. Interpretation: Nearly three-quarters of PLHIV in Eastern and Southern Africa experienced multidimensional poverty. Integrating HIV and poverty-reduction efforts, prioritising education, employment, clean energy, water, sanitation, housing, and assets is required. Including HIV indicators in poverty surveys, and research to accelerate joint progress are required. Funding: This study received no external funding Research in context: Evidence before this study: We searched PubMed, Google Scholar, reports by UNAIDS, UNDP, World Bank, and other grey literature, using subject headings and keyword terms ("HIV and poverty", "Differences in poverty between people living with HIV and people not living with HIV", "Multidimensional poverty and HIV", "HIV and sanitation", and "Differences between PLHIV and general population in assets") for English-language publications from January 1, 2005, to September 31, 2025. We reviewed 52 articles published in English (Supplement A1). The studies showed that people living with HIV (PLHIV) face socioeconomic disadvantage, including material deprivation, reduced employment, and limited household assets. The studies find as association between poverty among PLHIV with poorer immunologic and virologic response to antiretroviral therapy, lower adherence, and greater comorbidity. They further show inadequate access to safe drinking water, sanitation, and hygiene, increasing diarrhoeal disease and reducing treatment absorption among PLHIV. Other studies find that people living with HIV are deprived in cooking fuels, leading to upper respiratory infections. Most studies used single indicators of poverty or were restricted to individual settings or countries. Added value of this study: This study, to our knowledge, provides the first multi-country assessment of multidimensional poverty by HIV status in Eastern and Southern Africa. We find across countries, that people living with HIV (PLHIV) were more likely to be vulnerable to poverty, poor, or severely poor than people without HIV. Multidimensional poverty among PLHIV was driven by deprivations in education and employment, and deprivations in living standards such as electricity, clean cooking energy, safe drinking water, sanitation, housing, and household assets. Implications of all the available evidence: Despite major gains in HIV treatment and prevention and in national poverty reduction efforts, PLHIV continue to face socioeconomic and infrastructural disadvantage. These deprivations increase PLHIV’s risk to comorbidities and undermine HIV prevention and treatment services, slowing progress toward the Sustainable Development Goals. Integrating HIV responses with poverty reduction and social protection strategies is essential. Incorporating HIV indicators into national poverty monitoring systems and prioritising investment in education, employment, and essential services can accelerate joint progress towards ending AIDS and reducing poverty, in a context of declining external funding.

We assessed socioeconomic inequalities in social protection coverage among the public, men and women living with the human immunodeficiency virus (MLHIV, WLHIV), and adolescent girls and young women (AGYW). We used population-based data from Cameroon, Côte d’Ivoire, Ethiopia, Eswatini, Kenya, Lesotho, Malawi, Namibia, Rwanda, Tanzania, Uganda, Zambia, and Zimbabwe. We constructed concentration curves (CC) and computed concentration indices (CIX) for each country and population group. A CC represents the cumulative percentage of social protection coverage plotted on the y-axis against the cumulative proportion of the population—ranked by socioeconomic status from the poorest to the richest—on the x-axis. The CIX quantifies the concentration of social protection coverage among the poor or the rich. The sample size ranged from 10,197 in Eswatini to 29,577 in Tanzania. Social protection coverage among the public varied from 5.2% (95% Confidence Interval 4.5%–6.0%) in Ethiopia to 39.9% (37.0%–42.8%) in Eswatini. It ranged from 6.9% (5.7%–8.4%) MLHIV in Zambia to 45.0% (41.2–49.0) among WLHIV in Namibia. Among AGYW, it varied from 4.4% (3.6–5.3) in Ethiopia to 44.6% (40.8–48.5) in Eswatini. Socioeconomic inequalities in social protection coverage favored the poor in 11/13 countries surveyed. It favored the rich in Cameroon and was undefined in Côte d’Ivoire. The CIX in these 11 countries ranged from −0.080 (p = 0.002) among the public in Malawi to −0.372 (p< 0.001) among WLHIV in Zimbabwe. In 8 of these 11 countries, ≥15% of people from the poorest households reported receiving social protection. Only in countries with higher levels of social protection coverage did most people from the poorest households achieve high coverage. Social protection coverage was low and favored the poor. Pro-poor social protection is insufficient to reach the poor. Research is required to reach the poorest households with social protection in Africa.