The other end of healthspan: self-directed healthy years and avoidable delay

Vitalii Choban ORCID: 0009-0002-2770-9127, Independent Researcher, Varna, Bulgaria. Contact: [email protected].

Preprint. Not peer reviewed. Not conclusive, and not medical advice or a recommendation to any individual.

Version 2, 2026-11-15. License: CC BY 4.0.

Cite as: Choban, V. (2026). The other end of healthspan: Self-directed healthy years and avoidable delay (Version 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24243100

Record on Zenodo · PDF · This page as Markdown

Abstract

People are not born able to direct their own lives; that ability comes later. Delay in reaching it can cost years of healthy, self-directed life without shortening life at all. It is studied as part of the transition to adulthood and, where it is imposed, as waithood. It may come from economic hardship and other barriers, perhaps from family, or from a person’s own avoidance. Life extension research mostly tries to move the end of healthy life later; this paper looks at the other end. I call the years a person is both healthy and self-directed, meaning ready and realistically able to direct their own life, self-directed healthy years. Growing up faster is not the aim. Very early transitions such as early parenthood are associated with worse later health, lateness can reflect advantage, and the years before self-direction have value of their own. What matters is whether a delay is imposed or chosen, not how long it lasts. I argue that delay has a cost that health expectancy measures do not separately show, and that part of it comes from obstacles a person cannot remove alone. Because self-direction here requires being realistically able, such obstacles delay it by definition; how large that delay is, and whether they also make people less ready, are empirical questions. Removing such obstacles could make more of a life’s healthy years self-directed ones, without any biomedical discovery. That gain adds no years of life or health, so it complements geroscience and does not replace it. To make the argument testable, I define a population indicator, a measure of avoidable delay, and a break-even rule for preparation. Whether such delay can in practice be reduced, and whether reducing it would also improve later health, is untested, and I set out predictions that, if they fail, would show where the argument is wrong.

Keywords: self-directed healthy years; avoidable delay; transition to adulthood; emerging adulthood; waithood; healthspan; healthy life expectancy; life extension; public health.

1. Introduction

A 30-year-old has lived for 30 years, but not 30 years in which they could direct their own life. By this count, one who reached that point at 18 has had twelve such years, and one who reached it at 25 has had five. Without claiming that either age is better, this paper argues that the difference matters, that health expectancy measures do not show it, and that part of it may be avoidable.

Longevity research increasingly aims to extend the healthy part of life, not only its length. The gap between life expectancy and health-adjusted life expectancy, estimated at 9.6 years globally in 2019, is treated as a central problem (Garmany & Terzic, 2024). Most longevity research concentrates on the late end of life. As Kaeberlein (2018) notes, a common definition implies that healthspan is “a length of chronological time beginning at birth”. Counted from birth, there seems to be nothing to gain at the early end. But the early end does matter for one question: how long a person has to direct their own life in good health.

Since the later twentieth century, the transition to adulthood has lengthened in the United States (Hayford & Furstenberg, 2008). There, the share of 30-year-olds who had completed all five major role transitions (leaving home, finishing school, financial independence, marriage and parenthood) fell between 1960 and 2000 from 77% to 46% among women and from 65% to 31% among men. Without marriage and parenthood, the fall was much smaller: from nearly 85% to nearly 75% among women and from about 82% to 70% among men (Furstenberg et al., 2004). These trends count roles actually taken up, and later marriage and parenthood are not the delay this paper is about. Whether people have also become ready and able to direct their own lives later is unknown. The longer transition has its own literature in psychology and sociology (Arnett, 2000; Côté, 2000). A further literature, begun in the Middle East and Africa, describes delay imposed by a lack of work and of the means to form a household, and calls it waithood (Honwana, 2012; Singerman, 2007). To my knowledge, these have rarely been connected with health expectancy.

This paper reports no new data. It offers definitions, two measures, a break-even rule for preparation and a set of predictions that could fail. Part of the central claim holds by definition (Section 2.2), so what remains to be shown is how large avoidable delay is and whether reducing it would help. Where that delay comes from structural barriers, it overlaps with what the waithood literature calls involuntary waithood. This paper adds three things: a definition by ability instead of by roles; the other sources of delay; and a count in healthy years.

The evidence below is a selective narrative review, not a systematic one: targeted lookups and searches of PubMed, Europe PMC, OpenAlex and CORE between 2 and 6 November 2026, with a check of PsycINFO for work linking the transition to adulthood with health expectancy, where abstracts combining terms for the two returned no records. A search of OpenAlex, CORE, PubMed and Europe PMC for waithood on 12 November 2026 found no work that uses health expectancy.

2. What delay can cost

2.1 Self-direction and self-directed healthy years

No single criterion defines adulthood. Legal majority varies between 15 and 21 regardless of individual maturity (Knežević, 2018). No single role transition is necessary or sufficient for adult status (Hayford & Furstenberg, 2008), so treating marriage and parenthood as requirements would count many capable adults as not yet adult. In mainly American studies, young people themselves put accepting responsibility for oneself and making independent decisions first, with financial independence near the top (Arnett, 2000). Developmental studies offer no sharp line either: in cross-sectional data from eleven countries cognitive capacity reached adult levels around 16 and psychosocial maturity, in most of them, beyond 18 (Icenogle et al., 2019), and others argue that emerging adults are “intrinsically and physiologically, not yet adults” (Hochberg & Konner, 2020). If they are right, more of the delay is necessary.

I therefore separate three things. Readiness covers capacities, much of what developmental studies call psychosocial maturity: self-regulation, planning, judgement of consequences, and responsibility for oneself and towards others. It includes being able to act on one’s considered wishes despite distress, so disabling avoidance is a shortfall in readiness. It has to be measured by something other than the fact that the person remains dependent, or the reasoning would be circular. Opportunity is the realistic possibility of acting on those wishes, including living independently if the person chose to. What counts is what the person could do with the options open to them now, including a change in how they spend their time; options that further preparation would open later do not count, or everyone still in training would already count as able. What counts as living independently is a convention, which estimates should state and vary. Attainment covers roles actually taken up, such as leaving home or stable income. Self-direction here means readiness together with opportunity, and does not depend on attainment.

These follow distinctions in the capability approach. Readiness is what Nussbaum (2000) calls internal capabilities, “mature conditions of readiness”. Self-direction is a combined capability in her sense, “internal capabilities combined with suitable external conditions for the exercise of the function”, here the function of directing one’s own life. Attainment is functioning, what a person actually does.

Counting attainment would class someone who could leave home and chooses to stay as not self-directed, and would credit a 17-year-old forced to support a family with self-direction. A ready person who cannot afford to live independently is not yet self-directed in this technical sense, although nothing is lacking in them: the obstacle is outside the person. The same applies later in life: someone who stays at home to raise children, supported by a partner, remains self-directed while living independently stays a realistic option, and where it does not, what is missing is opportunity, not anything in them. Transition below means taking up adult roles.

Self-directed healthy years (SDHY) are the years a person is both healthy and self-directed. For one continuous period of health and self-direction:

SDHY = (age at end of healthspan) − (age at start of self-direction)

This form is a conceptual device, since health is not binary and there are “no accepted or validated metrics” for healthspan (Kaeberlein, 2018). Consider two people whose healthspan ends at 60, one self-directed from 25 and the other from 35. The first has 35 self-directed healthy years, 40% more than the second’s 25. Reaching the same gain from the other end would mean moving the end of healthspan to 70. If the second person’s delay was imposed, that gain was available without extending life. If it was chosen, for instance a longer course of study by someone who could not yet have lived independently without one, or was needed for recovery, it is not the delay this paper is about. A person who could have lived independently and chose to study was self-directed throughout, and only took up adult roles later.

These years are worth counting separately because of what they allow, although a healthy year spent not yet self-directed is not claimed to be worth less overall. What such a year lacks is the chance to direct one’s own life, and where the person would not endorse this, it is a loss in this one respect. This is close to the capability approach, which asks what people are actually able to do or to be rather than only how satisfied they are (Nussbaum, 2000; Sen, 1985).

2.2 Where delay comes from

Delay is avoidable when a person is ready, or could be with reasonable support, but a removable obstacle stands in the way. The test is ability: whether they could do otherwise if they wanted to, rather than what they say about why they stay. This cannot be settled for any one person, but it can be partly checked from outside (Section 5.2). Delay can arise from three directions, which often combine.

Structural barriers. Unaffordable housing, a lack of stable work, untreated mental health problems and unsupported disabilities can all keep dependent someone who is ready, or who could be with support. Where the barriers are economic, this overlaps with what is called waithood. The term was introduced for young people in the Middle East who remain unmarried and financially dependent on their families into their twenties and thirties while they save for the costs of marriage (Singerman, 2007). For Africa it has been described as “an involuntary delay in reaching adulthood”, in which young people cannot find work, marry or set up families of their own (Honwana, 2012). Later work separates waithood that is unintended from waithood that is planned, as when women postpone marriage for education and careers (Smith-Hefner & Inhorn, 2020). Waithood is marked by roles, above all marriage, which count here as attainment, so the two ideas overlap without coinciding. Involuntary waithood is one instance of avoidable delay where it leaves a ready person without the means to live independently; where a person with those means is waiting only to marry, it is not.

The quantitative evidence on these barriers comes from the United States and Britain. Across US states from 1960 to 2011, fewer jobs, lower wages and higher rents were each estimated to raise the share of young adults living with their parents. In that study’s simulation these together account for about 60% to 80% of the rise from 1970 to 2011 among men and 45% to 65% among women, depending on age, and the estimated effects changed little when marriage and school enrolment were controlled (Matsudaira, 2016). In British panel data, higher regional house prices slowed leaving home and encouraged returning, while a higher income of the young person’s own did the reverse (Ermisch, 1999). Neither study measures readiness. Because self-direction includes opportunity, barriers to opportunity delay it by definition; how large that delay is, and whether barriers also delay readiness, are empirical questions. Young people in mental health care, foster care or the justice system, and those with disabilities or chronic illness, often lose services abruptly during this transition, although few interventions had been tested by 2010 (Osgood et al., 2010). A later systematic review of 14 randomised or quasi-experimental studies of young people leaving care (all but one from the United States) found the evidence insufficient to recommend any approach (Taylor et al., 2024).

Family. Parents may contribute in more than one way. The route examined here is overprotection that withholds practice at making decisions. Overparenting has small, mostly cross-sectional associations with offspring anxiety and depression (r of about .16 and .20; Hu et al., 2025). But a systematic review found the causal evidence insufficient (Vigdal & Brønnick, 2022), a cross-lagged study of Chinese students found emotional problems preceding helicopter parenting rather than the reverse (Gao et al., 2023), and a preregistered Dutch study of 2,229 adolescents linked perceived overprotection to functioning mainly at the same time rather than over the following years (Arslan et al., 2023). None of these studies measured readiness itself. That overprotection can delay it is widely theorised but remains unproven.

Oneself. Choice needs care in both directions. A transition should be freely chosen, and waiting that a person endorses is not counted as a loss. Some evidence fits this: among 224 Belgian 22- and 23-year-olds, autonomous motivation for one’s living arrangement was more strongly related to wellbeing than the arrangement itself, although those living independently were more satisfied with it (Kins et al., 2009).

But saying “it is my choice” does not make it free, for at least five reasons. Fear: people avoid decisions by postponing them or doing nothing (Anderson, 2003), and clinical experience suggests that anxiety can sustain avoidance, with parents’ accommodation reinforcing it (Lebowitz, 2016), although failure to launch, a label for highly dependent adult children, remains little studied (Calentino et al., 2026). Pressure: a person may stay because parents object or because leaving would feel disloyal, motives that Kins et al. (2009) class as controlled. Resignation: desires can adapt to a person’s predicament (Sen, 1985), so someone stuck at home may stop wanting to leave, and a move home described as a choice may be a way of saving on housing in a downturn (Matsudaira, 2016). Inertia: people tend to prefer the status quo and inaction (Anderson, 2003), and staying may simply be comfortable, with no fear involved. Not knowing: someone may never have seen the alternatives, although I know of no direct evidence for this in the transition to adulthood. These do not rule out choice: a person can weigh a fear and still decide to stay.

The test of ability treats these differently. In their strong form, four of them mean the person could not do otherwise, and so count as delay: disabling fear, pressure close to coercion, resignation before a real barrier, and not knowing where the person had no realistic way to find out. Someone able to leave who stays through mild fear, ordinary guilt or inertia is not counted, but staying may still cost them years they would have wanted. In the long run people tend to regret inaction more than action, and such regrets more often come from an accumulated, unfocused pattern than from one decision (Gilovich & Medvec, 1995). Closing opportunities may raise the cost: older adults perceive fewer chances to undo regrets, and in cross-sectional data intense regret was associated with lower quality of life among older but not young adults (Wrosch et al., 2005). None of these findings is specific to this transition. The test is not duration: a long delay can be chosen, and I know of no evidence for a length of time that is too long. Where ability ends and choice begins is not settled, and from outside the two can look alike.

What is not counted as a loss. Delay is not avoidable when it is developmentally needed, is preparation the person needs before they could live independently, or is freely chosen. Schooling taken only because employers demand a higher credential has no clear status, and I leave it unresolved.

3. Why timing alone is not the test

This proposal does not concern accelerating biological development. The evidence does not support social speed as a goal, and it does not make lateness the problem either.

Early social transitions. A narrative review of mainly US work concluded that very early parenthood or not attending college seem to carry the most risk for poor health and wellbeing, but that much of this may reflect family background and earlier experience (Oesterle, 2013). An early role taken on because of adversity does not by itself show that the person is self-directed.

Education. Each additional year of education, up to 18, is associated with a 1.9% lower risk of adult mortality (95% uncertainty interval 1.8–2.0), although this is observational and shows publication bias (IHME-CHAIN Collaborators, 2024).

Who is late. Lateness can reflect advantage: in pooled panel data from Austria, Bulgaria and France, young adults from higher socioeconomic backgrounds were less likely to intend to marry or have a first child within three years, a difference found only before age 26, although more likely to intend to leave home (Billari et al., 2019).

Why they are late. The circumstances of delay seem to matter. Among 18- to 24-year-olds from one Ohio county, returning to the parental home, compared with never leaving it, was associated with more depressive symptoms only for those who gave losing or not finding a job as a reason for living there (Copp et al., 2017). In initial findings from a British cohort, the 1% of men and 2% of women not in education, employment or training throughout ages 16 to 24, who already had worse adolescent health, were more than three times as likely as those never in that position to report fair or poor health at 51, a gap that narrowed but remained once early-life disadvantage was taken into account (Pelikh et al., 2026).

Waithood. Small qualitative studies are consistent with this. All 12 young people interviewed in Zimbabwe described stress over money, which the authors tie to being stuck in waithood (Risinamhodzi & Mafa, 2025). In interviews with 38 young migrants in South Africa, the authors read waiting as leading them to treat risks such as unprotected and transactional sex as necessary for now, and argue that this raises HIV risk (Bernays et al., 2020). Neither study has a comparison group, so neither separates the effect of waiting from that of poverty.

The evidence on whether gaining independence itself helps is mixed. In a 27-year Norwegian cohort, leaving home was associated with a small drop in depressed mood, which did not differ by household income and, by parental education, was clear only in the highest group (Jørgensen et al., 2024). In a Seoul panel with up to four annual waves, which did not follow those who left Seoul, those moving into independence did not differ significantly in health trajectory from those who stayed with their parents (Park & Wu, 2026). In a British panel of 9,714 people aged 21 to 35, returning to the parental home was associated with slightly less mental distress, although more distressed people were also more likely to return (Wu & Grundy, 2023). In a US panel of people aged 18 to 27, losing a job in the Great Recession raised worry among those living independently but not among those living with their parents, whose distress rose instead when the father lost his job (Alam & Bose, 2022). Neither of these two studies separates people who stayed or returned by choice from those who were blocked.

For these reasons the analysis turns on ability and choice rather than on lateness.

4. A break-even rule for preparation

The age at which health ends may itself depend on how a person enters adulthood. Writing T(a) for that age given self-direction from age a, SDHY(a) = T(a) − a. If a year of preparation takes s years of self-directed time, it changes self-directed healthy years by ΔT − s, where ΔT is how far it moves the end of health. Here s is the time by which the preparation postpones the point at which the person could first live independently (Section 2.1). A year of training that postpones it by the whole year, because nothing open to the person in the meantime would let them live independently, has s = 1, so it must add more than one year to T. Someone studying at 24 who already runs their own life, or could if they chose to, is self-directed throughout, so s = 0 and the study costs no self-directed years. Waiting that is merely blocked prepares nothing, so this rule applies only to time spent preparing.

The break-even point (called minimum investment in version 1) is the level of preparation below which earlier self-direction would cost more healthy years later than it gains at the start. A single such point requires the net return, ΔT − s, to fall steadily, and it is then also the point of most self-directed healthy years: each further year adds less than s, so it costs self-directed healthy years, although where the person values that preparation it is not a loss in the sense of Section 2.2. If returns come in steps, as credentials might, it is better described by what preparation secures: skills that work requires, resources for a healthy life, and a way out of harmful work.

The rule holds looking back, for a person who lived to the end of their healthy life. In a population, and in expectation for a person deciding in advance, years lost at the start fall where nearly everyone is alive and years gained at the end where some have died, so unless preparation also delays death the bar is higher than s (Section 5.3). All years are counted as equal, without discounting.

The break-even rule concerns health returns and avoidable delay concerns barriers, so the two are separate questions. A person below the break-even point who wants to transition and is blocked is still blocked; the rule only warns that removing that barrier alone may not serve their health.

The existence and size of a break-even point are hypotheses. No study I found estimates how far more preparation before transition moves the age at which health ends. Studies of schooling, below, mostly measure mortality; where reform studies also measured later health, it changed little: self-rated health and limiting long-term illness in Britain (Clark & Royer, 2013), hospital stays and prescribed drugs in Sweden (Meghir et al., 2018), and hospital stays, diagnosed conditions, self-rated health and body mass index in West Germany (Hollenbach et al., 2026). A French compulsory schooling reform had mixed effects at ages 45 and over: better cognition in men, more depressive symptoms in women, and no change in physical functioning (Courtin et al., 2019). An observational study of early parenthood in women found that education, finances, depressive symptoms, smoking and obesity partly accounted for its link with later heart problems (Lee & Ryff, 2016). Effects of schooling on mortality exist in some contexts and not others (Galama et al., 2018).

If mortality rises exponentially with age at rate b, a hazard ratio r that lasts for life adds about ln(1/r)/b years of life. With b = 0.09, as in Section 5.3, two studies designed to isolate the effect of extra schooling imply between about zero and about half a year of life, not of health, per year of schooling: an upper bound of about 0.1 years from a British reform (r = 0.991; Clark & Royer, 2013) and about 0.5 years within Swedish twin pairs (r = 0.954; Lundborg et al., 2016). The observational association in Section 3 (r = 0.981) converts to about 0.2 years, an upper figure, because that association weakens with age, to 0.8% per year of schooling after age 70. A lower b would raise these figures in inverse proportion.

A Swedish reform that added about a quarter of a year of schooling had no detectable effect and rules out gains above 1 to 1.4 months of life within the ages observed, up to about 75 (Meghir et al., 2018); that bound is for a quarter-year of schooling and for deaths before 75, so it is not on the same scale and is left out of the comparison that follows. An older US study of compulsory schooling laws from 1915 to 1939 reports as much as 1.7 years of life per year of education at age 35 in 1960 from its OLS estimate, with a larger instrumental estimate that does not differ significantly from it (Lleras-Muney, 2005). Even if the end of health moved by as much as the end of life, the three converted figures fall short of one year and only that older estimate exceeds it, so on most of this evidence a year of schooling that took a full self-directed year (s = 1) would not be made up by its effect on mortality alone. This is not an argument against schooling: the count leaves out everything else education is worth. It does mean that the education evidence of Section 3 argues against speed only where preparation takes little self-directed time.

These studies use people moved by a school reform or differing within twin pairs, who may not be typical of those choosing later study. The British and Swedish reforms concern schooling at 14 to 16, which displaces almost no self-directed years, and they find about zero. The US laws were those in force when a person was 14, so that larger estimate also concerns schooling that displaces almost no self-directed years. The twin estimate reflects schooling beyond the compulsory years: among pairs who both had under ten years of schooling it was close to zero (Lundborg et al., 2016). So for later schooling, the margin that matters, the evidence here is one study, whose estimate also weakens with age. None of these estimates is in healthy years.

5. Making the argument testable

5.1 A population indicator

Sullivan’s method combines a life table with the age-specific prevalence of health (Saito et al., 2014; Sullivan, 1971). Let S(x) be survival to age x, h(x) the proportion of those alive who are healthy, and c(x) and p(x) the proportions of them who are self-directed and who are both. Then self-directed healthy life expectancy (SDHLE), per birth, is:

SDHLE = ∫ S(x) · p(x) dx

Survey data are needed for p(x), since health and self-direction are unlikely to be independent; p(x) ≈ h(x) · c(x) is an approximation whose error should be reported. Children who die before becoming self-directed contribute zero, so they stay in the statistic. The indicator partitions healthy life expectancy:

HLE = SDHLE + ∫ S(x) · [h(x) − p(x)] dx

The second term is healthy years not currently self-directed. It is not, as a whole, a loss: most of it is childhood and development, which have their own value. Section 5.2 separates the part that may be a loss.

A close precedent is healthy working life expectancy, the years a person aged 50 can expect to spend both healthy and in paid work (Lièvre et al., 2007; Parker et al., 2020). A joint indicator built on self-reports already captures access to opportunities such as jobs (Parker et al., 2020), so it is better for describing how many such years there are than for showing what produced them, and the same applies to SDHLE. Years of good life applies the Sullivan method to years above minimum thresholds for poverty, physical and cognitive health and life satisfaction, reported from ages 20 or 50 (Lutz et al., 2021). SDHLE differs from both in its state, and from healthy working life expectancy in covering the ages at which that state is entered, so the start can itself be studied.

Four choices need stating. Health is simplest to treat as yes-or-no for a joint prevalence, as in Eurostat’s activity-limitation measure (Eurostat, n.d.). This is a different measure from health-adjusted life expectancy, which lies behind the 9.6-year gap in Section 1 and weights years by the severity of health states; a yes-or-no measure ignores severity, and an arbitrary cutoff makes comparisons “extremely sensitive to slight differences in the cutoff used” (Salomon et al., 2012). Self-direction is also treated as yes-or-no so that it can enter a joint prevalence, although how much of their life a person can direct comes in degrees and can go on widening after the threshold is passed; a graded measure is left for later work. Method matters because self-direction can reverse, since opportunity can be lost, for instance in a recession: for healthy working life, ignoring return to the state may understate estimates (Lynch et al., 2022). Time matters because a period estimate combines the start of self-direction at about 25 with health at about 70 from cohorts 45 years apart. If timing is shifting, it mixes a change in when people become self-directed with a change in how many years they have, as period fertility rates mix tempo and quantum, so comparisons over time need cohort or multistate estimates. Results should be reported by sex and social background.

An existing survey could give a first rough estimate of when self-direction begins. One candidate is the Transition into Adulthood Supplement of the US Panel Study of Income Dynamics, which covers ages 18 to 28. Its 2017 questionnaire asks about health, living arrangements, earnings, savings and parents’ financial help (Panel Study of Income Dynamics, 2019). Its nearest items ask how much responsibility respondents take for earning their own living, paying rent and bills and managing money, and how often they feel nervous or shy with other people. I found in it no question on wanting to live independently and no measure of avoidance, so it would bound that start rather than estimate it.

5.2 Avoidable delay

SDHLE alone rewards earlier self-direction. A period estimate credits the early gain at once, while any health cost arrives decades later and is not attributed to it, and no form of SDHLE can tell imposed from chosen timing. The framework therefore needs a second concept. Let p*(x) be the joint prevalence expected if barriers were removed, with developmental needs and free choices left as they are, and S and h held fixed as a first approximation. Avoidable delay is:

AD = ∫ S(x) · [p*(x) − p(x)] dx

Removing a barrier to opportunity cannot lower p when readiness is held fixed, removing disabling avoidance (below) can only raise readiness, and with h fixed neither can raise p above h, so p(x) ≤ p*(x) ≤ h(x) at every age. This holds for barriers outside health, such as housing, work or family pressure, whose removal moves healthy years into self-directed ones. Where the barrier is itself a health problem, such as untreated mental illness or an unsupported disability, removing it can raise h as well, adding healthy years too, so with h fixed AD captures only part of that gain. The empirical questions are how large the gain is and what it costs. The integral runs over all ages, and self-direction can be lost again (Section 5.1), so AD also counts later losses of opportunity: a job lost in mid-life, financial dependence on a partner with no realistic way to live independently, or a healthy older person who cannot afford to live alone. These are losses rather than delay and may fall unevenly by sex, so they should be reported separately from delay before self-direction is first reached.

Avoidance counts as a removable obstacle here only where it is disabling, so that the person could not act on their wishes (Section 2.2), and measured independently of staying; where it is a symptom of a mental health problem, avoidance and that problem overlap and should not be counted twice. Ability can only be estimated: opportunity from income, savings, the work open to a person and available support against the local cost of living independently, and readiness from what a person has managed, practical tasks and measures of avoidance, although someone who has never had the chance shows nothing either way.

AD should be reported as a range, because p*(x) leaves free choices as they are and it is uncertain which stated choices are free. The uncertainty concerns people who have the means to live independently and stay. The low estimate takes every such stated choice as free: these people are already self-directed and outside AD. The high estimate takes as free only choices that are stable over time and autonomously motivated, so it is an upper bound: it also counts people held back by mild fear or guilt, who fall outside AD. The scale for such motivation may partly track living arrangements, since those living with their parents reported less autonomous motives (Kins et al., 2009).

Ready people who lack the means because of a barrier are counted whatever they say, since removing the barrier gives them the opportunity whether or not they would use it, and a stated choice to stay may be resignation (Section 2.2); the share of them who say they want to leave should be reported beside AD. People able to do otherwise who stay through mild fear, guilt or inertia fall outside AD, so whatever they lose is not counted by this indicator.

Unlike the idealised removal in p*(x), a real barrier change can also set some people back, and aggregate estimates show only the net effect; separating gains from losses needs individual data. p*(x) cannot be observed directly. Advantaged groups cannot serve as the benchmark, because their later family formation partly reflects socialised intentions and structural constraints that favour postponement (Billari et al., 2019), and the “free choices” it holds constant are themselves shaped by barriers. Better evidence would come from studies in which a specific barrier changes, and from people who are ready, lack the means to live independently, and say that they want to. The health measure must also not exclude the people concerned: if health means no activity limitation and h is fixed, disabled people have no AD by construction.

5.3 Illustration

Table 1 applies the indicator to a synthetic population to show how it behaves. All inputs are invented and the values do not estimate any real population.

Table 1. SDHLE under synthetic scenarios (years, per birth).
Scenario LE HLE SDHLE Change in SDHLE
Reference: self-direction from median age 28 82.58 68.42 40.48 0.00
Self-direction from median age 27 82.58 68.42 41.47 +0.99
Health loss one year later; self-direction from 28 82.58 69.15 41.20 +0.73
Health loss one year later; self-direction from 27 82.58 69.15 42.20 +1.72
Health loss and death one year later; self-direction from 28 83.58 69.42 41.47 +0.99

Note. Inputs: Gompertz hazard 0.00003 · exp(0.09x); health declines logistically (midpoint 72, scale 7); self-direction rises logistically (scale 3); health and self-direction independent: p(x) = h(x) · c(x). For death one year later, no one dies in the first year and the hazard at each later age is that of a year younger. Integrated over ages 0 to 120 in steps of 0.1 years. Changes are computed from unrounded values.

In this invented population, a little under half of all years lived are both healthy and self-directed, which shows how the indicator reads rather than what any real population has. Moving the start of self-direction by a year changes SDHLE by almost a full year, because nearly everyone here is alive and healthy at the ages when it begins, while life expectancy and healthy life expectancy stay the same. Moving health loss a year later adds only 0.73, because about a quarter of people have died by the ages at which health is lost; moving death a year later as well, as in the last row, brings it to +0.99. The two gains add up when both ends move (fourth row, +1.72), because the model does not let self-direction affect health; in real populations they may interact. If death does not move, a year of delay at the start therefore needs health loss to come about 1.36 years later to make up for it (0.99/0.73), which is the higher bar of Section 4. If the typical age of health loss is 65 to 80 instead of 72, it is about 1.2 to 1.7: the more people die before losing health, the higher it is. Delayed ageing also offers what this approach does not, additional life (Goldman et al., 2013).

6. Implications and limitations

If the argument holds, it points to removing obstacles rather than to accelerating young people, and it gives people a way to think about their own timing. It gives no support to pressuring individuals, making children act as adults, reducing protections for young people, or cutting access to education.

  • Untested. SDHLE has not been computed for any real population, and avoidable delay may be small.
  • Measurement. Psychosocial maturity, in a sample of male juvenile offenders, and personality have been measured over time (Monahan et al., 2013; Roberts et al., 2006), but I know of no validated instrument that measures readiness and opportunity separately in the same people. The nearest I found is a capability measure for adults, ICECAP-A, which asks in a single question how far a person is able to be independent, so it cannot separate readiness from opportunity (Al-Janabi et al., 2012). Any threshold is a convention.
  • Choice. Someone stuck at home may come to say it was their choice, and retrospect is unreliable, so the test of ability is only approximate and the range for avoidable delay does not settle who chose freely. Nussbaum (2000) argues that recognising that desires adapt to circumstances need not be unacceptably paternalistic, provided no single view of the good life is imposed (political liberalism) and the aim is what people are able to do (capabilities) rather than what they actually do (functionings). A readiness criterion, which discounts some stated choices, goes further than that and remains open to the charge.
  • Culture and stigma. A definition built on autonomy risks casting interdependent ways of life as deficient, and needs local validation. Waithood is the hard case: where norms expect young people to live with their parents until they marry (Smith-Hefner & Inhorn, 2020), a person with an income may still have no realistic way to live independently. The test of ability applies here person by person. Such a norm is a barrier for someone who would leave if they could and faces pressure close to coercion (Section 2.2), and a way of life for someone who shares it, so the same norm can count for one person and not for another. The quantitative evidence cited here comes mostly from North America, Europe and East Asia, and that on barriers from the United States and Britain alone. From the regions where waithood was first described I cite only two small qualitative studies. The measure is for populations and is no verdict on any person.
  • Goodhart’s law. The definition gives no credit for forced roles, but practical measures will lean on role-based proxies, so measured SDHLE could be raised by pushing people into adult roles early. It should therefore never be reported without avoidable delay, early-transition harms and health outcomes.
  • Affluence as a rival explanation. Adult activities among US adolescents declined from 1976 to 2016, and among 12th graders from 1993 to 2015 the decline appeared whether or not their fathers had attended college (Twenge & Park, 2019), so at those ages slowing may reflect affluence as much as barriers. Whether this extends to the twenties is untested.
  • Delay may cost little. Arnett (2000) suggests that exploration becomes more feasible and attractive when people can expect long lives, and the experiences it brings may be valuable in themselves.

7. Testable predictions

  1. Measurement. A short instrument for readiness and opportunity, extending existing measures of psychosocial maturity, proves reliable, valid and predictive across cultures. If it does not, SDHLE cannot be estimated.
  2. Variation. Within countries, higher local housing costs and youth unemployment are followed some years later by lower readiness. That withholding some roles, particularly work, could delay readiness is tentative and untested. Opportunity tracks these costs by construction, so the informative test is on readiness; if readiness does not move, the argument narrows to opportunity alone.
  3. Harm of being blocked. Among people of similar family background and with readiness and health measured beforehand, those blocked by a barrier measured independently of the outcome, such as local housing costs, have worse later mental and physical health than those not blocked and those delayed by choice. Housing costs should be compared within areas over time, so that stable differences between areas do not explain the result. Education is treated as a step on the path, so it is not controlled for. People in involuntary waithood (Section 2.2) are a first group in which to test this. If blocked people do not have worse later health, the health case for counting blocked years as a loss fails, and only the value claim remains.
  4. Intervention. Natural experiments or trials that reduce a specific barrier increase wanted independence among ready people, without lowering readiness or worsening education, health or wellbeing. If the gain comes only at the cost of lower readiness or worse education, health or wellbeing, the policy case fails. If stated choices to stay fall when a barrier is removed, some of those choices were not free.
  5. Endorsed and avoided delay. Compare three groups of similar background, with mental health, readiness and avoidance measured beforehand: avoidant people who transition late, equally avoidant people who transition earlier, and people late by a choice they endorse, measured with a scale of autonomous motivation (Kins et al., 2009). The first group has worse later health and wellbeing, and more regret looking back, than the other two. The same comparison applies to pressure, using that scale’s controlled motives. If there is no difference, the distinction should be dropped.
  6. Break-even point. Among people with similar background and early health, and with resources fixed in advance, there is a level of preparation up to which a further year adds more healthy years than the self-directed time s it takes, and beyond which it adds fewer. In a population the comparison is with the higher bar of Section 5.3. If there is no such level, the concept should be dropped.
  7. Refutation. Avoidable delay should be abandoned if no group with such delay can be identified. People in involuntary waithood are the first candidates, once readiness and means are measured and not inferred from roles. It should be treated as unimportant if, across scenarios set out in advance, it stays below a smallest gain worth caring about, also fixed in advance.

A raw association between later marriage or parenthood and better health would not count against the argument, because Section 3 already expects it from social background. The test is the comparison within similar backgrounds.

8. Conclusion

A person’s self-directed healthy years have a start as well as an end, and years can be lost at the start without any shortening of life. More self-directed healthy years may be reachable not only by delaying ageing but also by removing what keeps people dependent when they are ready, or could be with support: economic and other structural barriers, which overlap with what is elsewhere called involuntary waithood, perhaps family, or their own avoidance. For barriers to opportunity this follows from the definition; how much it would gain, and whether it also brings readiness forward, is untested. Self-directed healthy life expectancy, reported with avoidable delay, would let this be measured and tested. Removing barriers does not mean growing up faster: the evidence that timing can come too early may partly reflect who makes early transitions, but it is stronger so far than the evidence for the gains from removing barriers that this paper proposes. The aim is that people become self-directed when it suits them, prepared well enough to stay healthy, and are not held back by removable barriers or by their own avoidance.

Declarations

  • Funding: The author received no financial support for the research, authorship, and/or publication of this article.
  • Conflicts of Interest: None.
  • Data and code: No new empirical data were analysed. Table 1 and the figures in Sections 4 and 5.3, apart from Meghir et al. and Lleras-Muney, who state their own figures, can be reproduced with sdhle_scenarios.py, archived with this preprint at https://doi.org/10.5281/zenodo.24243100 under the MIT License.
  • Use of AI: During the preparation of this work the author used Claude Opus 5.5, Claude Fable 5.1 and Claude Fable 5.2 (Anthropic), Qwen3.8-Flash-Next (Qwen Team, Alibaba Group) and Gemini 4 Argon (Google) in order to assist with drafting and editing, literature search and reading of sources, review of the manuscript, and coding. All output produced with their assistance was critically reviewed and edited by the author, who takes responsibility for the content.

References

  • Alam, S. A., & Bose, B. (2022). Stepping into adulthood during a recession: Did job losses during the Great Recession impact health of young adults? Health Economics, 31(8), 1730–1751. https://doi.org/10.1002/hec.4535
  • Al-Janabi, H., Flynn, T. N., & Coast, J. (2012). Development of a self-report measure of capability wellbeing for adults: The ICECAP-A. Quality of Life Research, 21(1), 167–176. https://doi.org/10.1007/s11136-011-9927-2
  • Anderson, C. J. (2003). The psychology of doing nothing: Forms of decision avoidance result from reason and emotion. Psychological Bulletin, 129(1), 139–167. https://doi.org/10.1037/0033-2909.129.1.139
  • Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55(5), 469–480. https://doi.org/10.1037/0003-066X.55.5.469
  • Arslan, İ. B., Lucassen, N., Keijsers, L., & Stevens, G. W. J. M. (2023). When too much help is of no help: Mothers’ and fathers’ perceived overprotective behavior and (mal)adaptive functioning in adolescents. Journal of Youth and Adolescence, 52(5), 1010–1023. https://doi.org/10.1007/s10964-022-01723-0
  • Bernays, S., Lanyon, C., Dlamini, V., Ngwenya, N., & Seeley, J. (2020). Being young and on the move in South Africa: How ‘waithood’ exacerbates HIV risks and disrupts the success of current HIV prevention interventions. Vulnerable Children and Youth Studies, 15(4), 368–378. https://doi.org/10.1080/17450128.2020.1739359
  • Billari, F. C., Hiekel, N., & Liefbroer, A. C. (2019). The social stratification of choice in the transition to adulthood. European Sociological Review, 35(5), 599–615. https://doi.org/10.1093/esr/jcz025
  • Calentino, A. E., Lawhead, C., & Klein, D. N. (2026). Failure to launch (FTL): A narrative review and call for research. Clinical Child and Family Psychology Review. Advance online publication. https://doi.org/10.1007/s10567-026-00582-6
  • Clark, D., & Royer, H. (2013). The effect of education on adult mortality and health: Evidence from Britain. American Economic Review, 103(6), 2087–2120. https://doi.org/10.1257/aer.103.6.2087
  • Copp, J. E., Giordano, P. C., Longmore, M. A., & Manning, W. D. (2017). Living with parents and emerging adults’ depressive symptoms. Journal of Family Issues, 38(16), 2254–2276. https://doi.org/10.1177/0192513X15617797
  • Côté, J. E. (2000). Arrested adulthood: The changing nature of maturity and identity. New York University Press. https://doi.org/10.18574/nyu/9780814763995.001.0001
  • Courtin, E., Nafilyan, V., Glymour, M., Goldberg, M., Berr, C., Berkman, L. F., Zins, M., & Avendano, M. (2019). Long-term effects of compulsory schooling on physical, mental and cognitive ageing: A natural experiment. Journal of Epidemiology and Community Health, 73(4), 370–376. https://doi.org/10.1136/jech-2018-211746
  • Ermisch, J. (1999). Prices, parents, and young people’s household formation. Journal of Urban Economics, 45(1), 47–71. https://doi.org/10.1006/juec.1998.2083
  • Eurostat. (n.d.). Healthy life years statistics. Statistics Explained. Retrieved November 2, 2026, from https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Healthy_life_years_statistics
  • Furstenberg, F. F., Kennedy, S., McLoyd, V. C., Rumbaut, R. G., & Settersten, R. A. (2004). Growing up is harder to do. Contexts, 3(3), 33–41. https://doi.org/10.1525/ctx.2004.3.3.33
  • Galama, T. J., Lleras-Muney, A., & van Kippersluis, H. (2018). The effect of education on health and mortality: A review of experimental and quasi-experimental evidence. In Oxford Research Encyclopedia of Economics and Finance. Oxford University Press. https://doi.org/10.1093/acrefore/9780190625979.013.7
  • Gao, W., Hou, Y., Hao, S., & Yu, A. (2023). Helicopter parenting and emotional problems in Chinese emerging adults: Are there cross-lagged effects and the mediations of autonomy? Journal of Youth and Adolescence, 52(2), 393–405. https://doi.org/10.1007/s10964-022-01702-5
  • Garmany, A., & Terzic, A. (2024). Global healthspan-lifespan gaps among 183 World Health Organization member states. JAMA Network Open, 7(12), e2450241. https://doi.org/10.1001/jamanetworkopen.2024.50241
  • Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What, when, and why. Psychological Review, 102(2), 379–395. https://doi.org/10.1037/0033-295X.102.2.379
  • Goldman, D. P., Cutler, D., Rowe, J. W., Michaud, P.-C., Sullivan, J., Peneva, D., & Olshansky, S. J. (2013). Substantial health and economic returns from delayed aging may warrant a new focus for medical research. Health Affairs, 32(10), 1698–1705. https://doi.org/10.1377/hlthaff.2013.0052
  • Hayford, S. R., & Furstenberg, F. F. (2008). Delayed adulthood, delayed desistance? Trends in the age distribution of problem behaviors. Journal of Research on Adolescence, 18(2), 285–304. https://doi.org/10.1111/j.1532-7795.2008.00561.x
  • Hochberg, Z., & Konner, M. (2020). Emerging adulthood, a pre-adult life-history stage. Frontiers in Endocrinology, 10, 918. https://doi.org/10.3389/fendo.2019.00918
  • Hollenbach, J., Schmitz, H., & Tawiah, B. B. (2026). Life-cycle health effects of compulsory schooling. The European Journal of Health Economics, 27(6), 1293–1322. https://doi.org/10.1007/s10198-025-01884-2
  • Honwana, A. M. (2012). The time of youth: Work, social change, and politics in Africa. Kumarian Press. https://doi.org/10.1515/9781565494732
  • Hu, N., Chen, K., Ye, L., Liu, H., Cai, D., Zhang, H., & Zhao, Y. (2025). Associations between overparenting and offspring’s mental health: A meta-analysis of multiple moderators. Behavioral Sciences, 15(9), 1235. https://doi.org/10.3390/bs15091235
  • Icenogle, G., Steinberg, L., Duell, N., Chein, J., Chang, L., Chaudhary, N., Di Giunta, L., Dodge, K. A., Fanti, K. A., Lansford, J. E., Oburu, P., Pastorelli, C., Skinner, A. T., Sorbring, E., Tapanya, S., Uribe Tirado, L. M., Alampay, L. P., Al-Hassan, S. M., Takash, H. M. S., & Bacchini, D. (2019). Adolescents’ cognitive capacity reaches adult levels prior to their psychosocial maturity: Evidence for a “maturity gap” in a multinational, cross-sectional sample. Law and Human Behavior, 43(1), 69–85. https://doi.org/10.1037/lhb0000315
  • IHME-CHAIN Collaborators. (2024). Effects of education on adult mortality: A global systematic review and meta-analysis. The Lancet Public Health, 9(3), e155–e165. https://doi.org/10.1016/S2468-2667(23)00306-7
  • Jørgensen, M., Smith, O. R. F., Wold, B., & Haug, E. (2024). Social inequality in the association between life transitions into adulthood and depressed mood: A 27-year longitudinal study. Frontiers in Public Health, 12, 1286554. https://doi.org/10.3389/fpubh.2024.1286554
  • Kaeberlein, M. (2018). How healthy is the healthspan concept? GeroScience, 40(4), 361–364. https://doi.org/10.1007/s11357-018-0036-9
  • Kins, E., Beyers, W., Soenens, B., & Vansteenkiste, M. (2009). Patterns of home leaving and subjective well-being in emerging adulthood: The role of motivational processes and parental autonomy support. Developmental Psychology, 45(5), 1416–1429. https://doi.org/10.1037/a0015580
  • Knežević, M. (2018). When do we become adults? Review of theory, research and recent advances from an interdisciplinary perspective. Psihologijske teme, 27(2), 267–289. https://doi.org/10.31820/pt.27.2.7
  • Lebowitz, E. R. (2016). “Failure to launch”: Shaping intervention for highly dependent adult children. Journal of the American Academy of Child & Adolescent Psychiatry, 55(2), 89–90. https://doi.org/10.1016/j.jaac.2015.10.014
  • Lee, C., & Ryff, C. D. (2016). Early parenthood as a link between childhood disadvantage and adult heart problems: A gender-based approach. Social Science & Medicine, 171, 58–66. https://doi.org/10.1016/j.socscimed.2016.10.028
  • Lièvre, A., Jusot, F., Barnay, T., Sermet, C., Brouard, N., Robine, J.-M., Brieu, M.-A., & Forette, F. (2007). Healthy working life expectancies at age 50 in Europe: A new indicator. The Journal of Nutrition, Health & Aging, 11(6), 508–514. https://pubmed.ncbi.nlm.nih.gov/17985068/
  • Lleras-Muney, A. (2005). The relationship between education and adult mortality in the United States. Review of Economic Studies, 72(1), 189–221. https://doi.org/10.1111/0034-6527.00329
  • Lundborg, P., Lyttkens, C. H., & Nystedt, P. (2016). The effect of schooling on mortality: New evidence from 50,000 Swedish twins. Demography, 53(4), 1135–1168. https://doi.org/10.1007/s13524-016-0489-3
  • Lutz, W., Striessnig, E., Dimitrova, A., Ghislandi, S., Lijadi, A., Reiter, C., Spitzer, S., & Yildiz, D. (2021). Years of good life is a well-being indicator designed to serve research on sustainability. Proceedings of the National Academy of Sciences, 118(12), e1907351118. https://doi.org/10.1073/pnas.1907351118
  • Lynch, M., Bucknall, M., Jagger, C., & Wilkie, R. (2022). Projections of healthy working life expectancy in England to the year 2035. Nature Aging, 2(1), 13–18. https://doi.org/10.1038/s43587-021-00161-0
  • Matsudaira, J. D. (2016). Economic conditions and the living arrangements of young adults: 1960 to 2011. Journal of Population Economics, 29(1), 167–195. https://doi.org/10.1007/s00148-015-0555-y
  • Meghir, C., Palme, M., & Simeonova, E. (2018). Education and mortality: Evidence from a social experiment. American Economic Journal: Applied Economics, 10(2), 234–256. https://doi.org/10.1257/app.20150365
  • Monahan, K. C., Steinberg, L., Cauffman, E., & Mulvey, E. P. (2013). Psychosocial (im)maturity from adolescence to early adulthood: Distinguishing between adolescence-limited and persisting antisocial behavior. Development and Psychopathology, 25(4 Pt 1), 1093–1105. https://doi.org/10.1017/S0954579413000394
  • Nussbaum, M. C. (2000). Women and human development: The capabilities approach. Cambridge University Press. https://doi.org/10.1017/CBO9780511841286
  • Oesterle, S. (2013). Pathways to young adulthood and preventive interventions targeting young adults. In Improving the health, safety, and well-being of young adults: Workshop summary (pp. 147–176). National Academies Press. https://doi.org/10.17226/18340
  • Osgood, D. W., Foster, E. M., & Courtney, M. E. (2010). Vulnerable populations and the transition to adulthood. The Future of Children, 20(1), 209–229. https://doi.org/10.1353/foc.0.0047
  • Panel Study of Income Dynamics. (2019). Transition into Adulthood Supplement, wave 7, 2017: Questionnaire. Institute for Social Research, University of Michigan. https://psidonline.isr.umich.edu/TAS/questionnaires/ta17.pdf
  • Park, G.-R., & Wu, J. (2026). Boomerang young adults in Seoul, South Korea: Residential pathways, health and social attitudes. Population, Space and Place, 32(7), e70360. https://doi.org/10.1002/psp.70360
  • Parker, M., Bucknall, M., Jagger, C., & Wilkie, R. (2020). Extending working lives: A systematic review of healthy working life expectancy at age 50. Social Indicators Research, 150(1), 337–350. https://doi.org/10.1007/s11205-020-02302-1
  • Pelikh, A., Parsons, S., & Crawford, C. (2026). Long-term consequences of being NEET in early adulthood: Initial findings from the 1970 British Cohort Study at age 51. UCL Centre for Longitudinal Studies. https://cls.ucl.ac.uk/wp-content/uploads/2017/02/Long-term-consequences-of-being-NEET-Initial-findings-from-the-1970-British-Cohort-Study-at-age-51.pdf
  • Risinamhodzi, N., & Mafa, I. H. (2025). ‘Stuck in waithood’: Factors underpinning stress-related challenges among youths in Zimbabwe. Journal of Humanities, 33(1), 68–90. https://doi.org/10.4314/jh.v33i1.4i
  • Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course: A meta-analysis of longitudinal studies. Psychological Bulletin, 132(1), 1–25. https://doi.org/10.1037/0033-2909.132.1.1
  • Saito, Y., Robine, J.-M., & Crimmins, E. M. (2014). The methods and materials of health expectancy. Statistical Journal of the IAOS, 30(3), 209–223. https://doi.org/10.3233/SJI-140840
  • Salomon, J. A., Wang, H., Freeman, M. K., Vos, T., Flaxman, A. D., Lopez, A. D., & Murray, C. J. L. (2012). Healthy life expectancy for 187 countries, 1990–2010: A systematic analysis for the Global Burden Disease Study 2010. The Lancet, 380(9859), 2144–2162. https://doi.org/10.1016/S0140-6736(12)61690-0
  • Sen, A. (1985). Well-being, agency and freedom: The Dewey Lectures 1984. The Journal of Philosophy, 82(4), 169–221. https://doi.org/10.2307/2026184
  • Singerman, D. (2007). The economic imperatives of marriage: Emerging practices and identities among youth in the Middle East (Middle East Youth Initiative Working Paper No. 6). Wolfensohn Center for Development and Dubai School of Government. https://doi.org/10.2139/ssrn.1087433
  • Smith-Hefner, N. J., & Inhorn, M. C. (2020). Introduction: Waithood: Gender, education, and global delays in marriage and childbearing. In M. C. Inhorn & N. J. Smith-Hefner (Eds.), Waithood: Gender, education, and global delays in marriage and childbearing (pp. 1–28). Berghahn Books. https://doi.org/10.1515/9781789209006-003
  • Sullivan, D. F. (1971). A single index of mortality and morbidity. HSMHA Health Reports, 86(4), 347–354. https://doi.org/10.2307/4594169
  • Taylor, D., Albers, B., Mann, G., Lewis, J., Taylor, R., Mendes, P., Macdonald, G., & Shlonsky, A. (2024). Systematic review and meta-analysis of policies and interventions that improve health, psychosocial, and economic outcomes for young people leaving the out-of-home care system. Trauma, Violence, & Abuse, 25(5), 3534–3554. https://doi.org/10.1177/15248380241253041
  • Twenge, J. M., & Park, H. (2019). The decline in adult activities among U.S. adolescents, 1976–2016. Child Development, 90(2), 638–654. https://doi.org/10.1111/cdev.12930
  • Vigdal, J. S., & Brønnick, K. K. (2022). A systematic review of “helicopter parenting” and its relationship with anxiety and depression. Frontiers in Psychology, 13, 872981. https://doi.org/10.3389/fpsyg.2022.872981
  • Wrosch, C., Bauer, I., & Scheier, M. F. (2005). Regret and quality of life across the adult life span: The influence of disengagement and available future goals. Psychology and Aging, 20(4), 657–670. https://doi.org/10.1037/0882-7974.20.4.657
  • Wu, J., & Grundy, E. (2023). ‘Boomerang’ moves and young adults’ mental well-being in the United Kingdom. Advances in Life Course Research, 56, 100531. https://doi.org/10.1016/j.alcr.2023.100531