Main results

The Florence research unit, based at the Department of Statistics, Computer Science, Applications “G. Parenti”, could count on only two researchers throughout the period: Gustavo De Santis and Mauro Maltagliati. We repeatedly tried to recruit a new researcher through a Research Grant (Assegno di Ricerca): in 2023, 2024 and early 2025, but failed. The first time (2023) because the winner withdrew after only three months, the second time (2024) because he withdrew immediately after being selected, and the third time (later in 2024) because nobody applied.

Despite this unexpected outcome, the research was successfully carried out along three lines:

a) an analysis of the geographical disparities in mortality at the micro level in Italy and their impact on mortality trends (the main thread of our research plan), and three side lines of research

b) health conditions in Italy (including causes of death)

c) the finding that, in shaping the population age structure in the long run, mortality, rather than fertility (as commonly believed), is the main driving force, and, in connection with this,

d) how to incorporate this finding into the design of a viable pension system.

With regard to point (a), we started from the analysis of the stall (if not reversal) of mortality convergence across Italian regions in the pre-Covid-19 years (Carboni et al., 2024), and investigated this phenomenon at the finest possible level, namely municipalities, using an ingenious standardisation technique to overcome data limitations, because available data were too sparse and raw (i.e. without the necessary age breakdown) to be analysed in the traditional way (e.g. by constructing life tables). We covered the pre-Covid-19 years (2002–2018) and were able to detect a clear disadvantage for the “frailest” municipalities, where frailty could be identified thanks to the combination of various municipal indicators prepared by Istat for the period under consideration (De Santis et al., 2026). To our surprise, we discovered that these “frail” municipalities cannot be ordered along the traditional North–South gradient. Rather, they can be better understood with reference to the more elusive notion of “marginal” areas, i.e. areas with poor infrastructures (e.g. in terms of schools and hospitals) and limited connections to municipalities where these services are available. Variability in survival appeared to be, all in all, limited in the period under consideration, with only a very modest tendency to increase. In short, the main concern that originated the research project (heterogeneity in survival chances on the increase in Italy?) found a non-conclusive answer: what we found is consistent with both the optimistic view (geographic heterogeneity in survival is low in our country and not increasing) and its opposite (no progress observed in the 17 years under scrutiny and, if anything, a very small tendency to increase).

With regard to point (b), we cooperated with Istat researchers in the preparation of a report on health and survival in Italy in recent years, and we found that the same uncertainty persisted in this case as well: while standards are generally high, also in terms of health conditions, heterogeneity is significant, and the most recent years (in this case, after the end of the Covid-19 pandemic) show some worrying signs of a possible worsening of conditions (De Santis et al., 2025).

With regard to point (c), an intuition, derived from the analysis of life tables and population data, led us to look more deeply into the matter and to conclude that, in the long run, and contrary to what most demographers think, mortality may be the main driving force behind population age structures (De Santis and Salinari, 2024). The main reason, in the simplest possible terms, is that mortality conditions, beyond affecting the age structure directly, also influence fertility, and this, in turn, impacts the age structure. Scholars have generally focused on the latter passage (from fertility to the age structure), but have overlooked the former (from mortality to fertility).

With regard to point (d), if mortality is the main driver of population age structures in the long run, it seems logical to take this explicitly into account in the construction of a well-designed, viable pay-as-you-go pension system, although, obviously, in this case mortality is only part of the story (De Santis, 2024).

All of these research threads were presented at conferences and seminars before their publication in scientific papers, and we benefited from discussions with colleagues: see section “Participation in conferences”.

We did not just pursue scientific analysis: we also made sure to disseminate our findings through some popular, but scientifically sound blogs, such as Neodemos (https://www.neodemos.info/), in Italian, and N-IUSSP (https://www.niussp.org/) and Population Europe (https://population-europe.eu/) in English: see section “Dissemination”.

The Sapienza Unit contributed to significant advancements in the comprehension of mortality dynamics over time by sex, space and causes of death. A first line of research focused on the analysis of the role of a group of emerging causes of death, the so-called “Deaths of Despair”, in Italy, at both national and sub-national level, in the overall trend of mortality. “Deaths of Despair” refer to mortality due to alcohol consumption, drug use and suicide. Results show that, although the Italian experience appears less severe than that observed in the US or other contexts, the impact remains non-negligible: in 2018, deaths of despair accounted for a cumulative loss in life expectancy of nearly five months (Lanfiuti Baldi et al., 2025; 2026).

Another important part of our research activity was devoted to methodological and technical issues, in particular to the distortion of conventional longevity measures due to tempo effects. We estimated the impact of tempo effects and computed tempo-adjusted life expectancy by sex at national and provincial level, and found that the impact was particularly significant during short-term mortality crises, when the observed drop in conventional life expectancy is driven by a temporary mortality shift rather than by a true deterioration in survival conditions. The analysis revealed systematic and spatially structured differences between conventional and adjusted estimates of life expectancy. This study contributes to a correct interpretation of longevity trends, which is essential for targeted policy-making. We are finalising a manuscript that will first be presented at the forthcoming annual conference of the Society of Economics, Demography and Statistics (SIEDS) in June 2026.

The Sapienza team also worked on the development of statistical models for the estimation of mortality risks, with a focus on old age. A manuscript based on this study is also being prepared for submission.

A further line of research concerned geographical differences in longevity. We mapped Italian municipalities according to two interpretations and measures of longevity, i.e. areas with a high concentration of long-lived individuals and areas with high life expectancy. We highlighted the existence of a few areas with a higher concentration of centenarians but with a definitely lower level of life expectancy, and these areas are all inner areas. This suggests that factors other than the efficiency of the health services offered must play a role in determining individual longevity in these apparently disadvantaged areas. A paper has been submitted for publication.

The project’s works in progress and results have been widely disseminated during the project through contributed and invited talks, and posters at major international demographic conferences: the Annual Meeting of the Population Association of America (PAA), the Population Days of the Italian Association of Population Studies (AISP), and the Meeting of the Italian Statistical Society (SIS). Preliminary results were also presented at the intermediate project workshop “Advances in causes of death modelling”, within the CARONTE PRIN–MIUR 2022 Grant: “Causes of deAth dependence stRuctures and the cOmpositioNal effecT on ovErall mortality”.

It is worth underlying that the Sapienza Unit encountered some difficulties in the formation of its research team. The first challenge arose when a researcher (RTDA), originally included as a participant, moved to another institution prior to the official project start date (October 2023). Furthermore, we launched a call for a Research Grant (Assegno di Ricerca) twice. Unfortunately, both calls yielded negative results, either due to a lack of suitable applicants or because candidates did not meet the specific technical requirements of the project. These combined factors (the early loss of a permanent researcher and the unsuccessful recruitment cycles) hindered the initial consolidation of the research group. Luckily, however, in the final phase of the project we successfully awarded a research contract, recruiting a highly qualified junior researcher whose expertise fully aligns with the project’s requirements.

As mentioned above, in addition to the publications already completed or currently being finalised, several research works initiated within the framework of this project remain ongoing. These studies, which further develop the project’s contributions in areas such as the impact of tempo effects on cause-specific mortality and the dynamics of mortality improvements at the population and individual level, will result in manuscripts to be submitted in the near future to high-ranking international demographic journals. The MUR will be acknowledged as the funding source even if the publication occurs after the end of the project timeline.

The Sassari research unit, based at the Department of Economics and Business Sciences of the University of Sassari, is currently composed of Giambattista Salinari (Associated Principal Investigator of the local unit), Silvia Battino, and Gianni Carboni. During the development of the project, the composition of the team changed. Massimo Esposito, who was initially part of the research unit, left the Department of Economics of the University of Sassari during the course of the project and therefore ceased to participate in the PRIN activities. Gianni Carboni joined the project in April 2024 as a research fellow. His fellowship, funded partly through the PRIN project and partly through a cascade PNRR project, will end in June.

The main objective of the project was to investigate the causes of the slowdown in the progression of life expectancy observed in Italy, a phenomenon that has also been documented in many other advanced economies characterised by low mortality levels. The initial idea for the project partly originated from a paper by the head of the Sassari unit (Salinari, Benassi and Carboni, 2023), which was under review when the project proposal was written and was subsequently published in the demographic journal Population Research and Policy Review. Although this article does not formally belong to the set of papers produced within the PRIN project, it can be considered the starting point of the research agenda developed by the Sassari group. The article proposes the hypothesis that the slowdown in life expectancy in Italy may be related to austerity policies in the healthcare sector introduced around 2008: after that year, and for more than a decade, net consolidated public health expenditure in Italy remained essentially stagnant. Over the following two years, members of the Sassari unit actively participated in numerous academic conferences, presenting research related to the themes of the PRIN project and publishing several scientific papers.

One of them (Carboni et al., 2024), published in Genus, focuses on the interruption of the convergence process in life expectancy across Italian regions. The study identifies the reforms of the Italian healthcare system implemented during the 1990s and early 2000s as the most plausible explanation for this break in the convergence process.

Two additional studies (Carboni, 2023, published in RIEDS, and Carboni et al., currently under review in Quality & Quantity, second round of revisions) investigate the relationship between economic crises and mortality. These papers identify a direct positive effect of economic downturns on mortality outcomes, a phenomenon sometimes referred to as the Thomas effect. In particular, the findings suggest that economic recessions are associated with improvements in mortality and reductions in lifespan inequality. These results imply that the direct economic consequences of the crisis (such as unemployment or impoverishment) cannot account for the observed slowdown in the improvement of life expectancy. It is worth noting that Carboni (2023) received the Best Young Researcher Award at the LIX Scientific Meeting of SIEDS, held in Naples.

Two further papers produced by the group (Carboni, 2026; Carboni and Salinari, 2026), both published in RIEDS, examine the statistical characteristics of the deceleration process in life expectancy. The results indicate that the slowdown is more consistent with a period effect (that is, a shock affecting the entire population starting around 2008) rather than with a cohort effect. This finding is particularly relevant because many alternative explanations of the slowdown in life expectancy implicitly or explicitly rely on cohort-based mechanisms, such as those related to the long-term effects of the obesity epidemic or the hypothesis that human populations may be approaching a biological limit to longevity. The analyses further show that age-specific mortality rates also exhibit a slowdown in their rate of improvement starting from the same period in which the deceleration of life expectancy becomes visible.

Naturally, the duration of the project (two years) represents a relatively short time frame for a research agenda of this scope, and part of the work initiated within the project is still ongoing. In particular, the research unit is currently completing an additional paper devoted to the analysis of the so-called “recovery plans” (piani di rientro) introduced in some Italian regions as part of healthcare austerity policies. This study focuses on a specific institutional mechanism through which fiscal constraints affected regional healthcare systems and is expected to be completed in the near future.

In addition to academic publications and conference presentations, the Sassari research unit has also been active in the dissemination of its results to a broader audience beyond the community of specialists. Members of the group contributed to public outreach and scientific communication through platforms such as Neodemos and N-IUSSP. Dissemination activities also included participation in the Italian Population Report published by the Italian Association for Population Studies (AISP).