Reflection on modern methods: risk ratio regression-simple concept yet complex computation

The risk ratio (RR) is the ratio of the outcome among the exposed to risk of the outcome among the unexposed. This is a simple concept, which makes one wonder why it has not gained the same popularity as the odds ratio. Using logistic regression to estimate the odds ratio is quite common in epidemiology and interpreting the odds ratio as a risk ratio, under the assumption that the outcome is rare, is also common. On one hand, estimating the odds ratio is simple but interpreting it is hard. On the other, estimating the risk ratio is challenging but its interpretation is straightforward. Issues with estimating risk ratio still remain after four decades. These issues include convergence of the algorithm, the choice of regression specification (e.g. log-binomial, Poisson) and many more. Various new computational methods are available which help overcome the issue of convergence and provide doubly robust estimates of RR.

Effects of environmental impact labels on the sustainability of food purchases: Two randomised controlled trials in an experimental online supermarket

Providing consumers with product-specific environmental impact information for food products (ecolabels) may promote more sustainable purchasing, needed to meet global environmental targets. Two UK studies investigated the effectiveness of different ecolabels using an experimental online supermarket platform. Study 1 (N = 1051 participants) compared three labels against control (no label), while Study 2 (N = 4979) tested four designs against control. Study 1 found significant reductions in the environmental impact score (EIS) for all labels compared to control (labels presented: values for four environmental indicators [-3.9 percentiles, 95%CIs: -5.2,-2.6]; a composite score [taking values from A to E; -3.9, 95%CIs: -5.2,-2.5]; or both together [-3.2, 95%CIs: -4.5,-1.9]). Study 2 showed significant reductions in EIS compared to control for A-E labels [-2.3, 95%CIs: -3.0,-1.5], coloured globes with A-E scores [-3.2, 95%CIs:-3.9,-2.4], and red globes highlighting ’worse’ products [-3.2, 95%CIs:-3.9,-2.5]. There was no evidence that green globes highlighting ’better’ products were effective [-0.5, 95%CIs:-1.3,0.2]. Providing ecolabels is a promising intervention to promote the selection of more sustainable products.

Causal inference and effect estimation using observational data

Observational studies aiming to estimate causal effects often rely on conceptual frameworks that are unfamiliar to many researchers and practitioners. We provide a clear, structured overview of key concepts and terms, intended as a starting point for readers unfamiliar with the causal inference literature. First, we introduce theoretical frameworks underlying causal effect estimation methods: the counterfactual theory of causation, the potential outcomes framework, structural equations and directed acyclic graphs. Second, we define the most common causal effect estimands, and the issues of effect measure modification, interaction and mediation (direct and indirect effects). Third, we define the assumptions required to estimate causal effects: exchangeability, positivity, consistency and non-interference. Fourth, we define and explain biases that arise when attempting to estimate causal effects, including confounding, collider bias, selection bias and measurement bias. Finally, we describe common methods and study designs for causal effect estimation, including covariate adjustment, G-methods and natural experiment methods.

Using linked administrative data to move from an individual unit to family units: An example from child protection policy to better inform family focused interventions

ObjectiveTo describe turning an incident based child protection system contacts and data (e.g. number of notifications) into child specific experiences and then into patterns of family experiences with child protection (CP) to inform family focused interventions and policy. ApproachData was drawn from Better Evidence Better Outcomes Linked Data (BEBOLD), a whole-of-population linked de-identified administrative data platform for all children in South Australia born from 1991-2016 (n~500,000) as well as their families and carers. Data linked included birth registrations, perinatal statistics, child protection services, family health services, education, justice, hospitalisation, and welfare payments. Families have been defined in BEBOLD based on mother-child relationships identified via birth registrations. The mother is the index family member, and all children registered with the same mother are grouped into one family. We analysed family experiences of the child protection system over ten years from 2006 to the end of 2016 and described their sociodemographic characteristics. ResultsAlmost 1 in 4 families had contact with CP 10 years after the birth of their first child. Two-thirds of families in contact with CP had one or two children, but larger families are more likely to have contact (57.1% of families with five or more children compared to 21.9% for families with one child).  Of the families that had CP contact, 56.5% had contact when they only had one child, 32.9% first had contact after birth of their second child, and 9% had contact had birth of third child. Families with only one child at first CP contact had a generally higher risk profile across a range of characteristics at birth than families who first had contact with two, or three or more children. ConclusionUnderstanding CP contact for family units more closely reflects how services seek to address CP risk by working with families. Describing how families interact with CP and identifying families at high risk of CP involvement can inform how agencies design and deliver services to target the most vulnerable families.

Intergenerational welfare and contact with the child protection system

ObjectivesSocioeconomic disadvantage and poverty are considered important drivers of child maltreatment, particularly child neglect. We examine the prevalence, cumulative incidence, and notification rate of child protection (CP) contact among children who experienced intergenerational welfare contact (IWC) compared to families with other types of welfare contact (WC). ApproachA whole-population linked administrative data study of children in South Australia born 1991-1995 (n=94,358), followed from birth up to age 20 years and their parent/s (n=143,814). The study used de-identified data from the Better Evidence Better Outcomes Linked Data (BEBOLD) platform including data from Australian Government Centrelink (welfare payments), birth registration, perinatal birth records, and child protection records. Children were classified as: No WC, parent only WC, child only WC, or IWC. Child protection contact prevalence, rates, and cumulative incidence were estimated by age, level of contact (reported notifications, screened-in reports, investigations, substantiations and out-of-home care), and WC group. ResultsUsing Australian Government Centrelink data, WC was defined as parent/s receiving a means-tested welfare payment (low-income, unemployment, disability or caring) when children were aged 11-15, or children receiving payment at ages 16-20. IWC was WC occurring in both parent and child generations. Children who experienced IWC accounted for 84.7% of all CP notifications at ages 0-17 years and had the highest prevalence of CP contact (37.3%, n=12,286), compared to children experiencing parent only WC (16.1%), child only WC (11.6%), and no WC (4.7%). This trend was consistent across all levels of CP contact. Children who experienced IWC had higher notification rates at every age compared to all other WC types. Conclusions and RelevanceChildren experiencing IWC represent a third of the population aged 11-20, yet account for 84.7% of all child protection notifications. Our study shows that children who experience IWC have higher personal (cumulative incidence and prevalence) and societal (notification rates) child protection costs than their counterfactual (parent only WC).

The hospital burden of intergenerational contact with the welfare system: A whole-of-population liked data study

ObjectivesIntergenerational welfare contact (IWC) is a policy issue due to personal and social costs of ‘entrenched disadvantage’ across generations, yet few studies have quantified the burden of IWC on health. We examine the proportion of children who experienced IWC and other welfare contact (WC) types; and estimate cause-specific hospital burden. ApproachA whole-of-population linked administrative data study of children in South Australia born 1991-1995 (n=94,358), followed from birth up to age 20 years and their parent/s (n=143,814). The study used de-identified data from the Better Evidence Better Outcomes Linked Data platform including data from Australian Government Centrelink (welfare payments), birth registration, perinatal birth records, and inpatient hospitalisations. Children were classified as: No WC, parent only WC, child only WC, or IWC. Cause-specific hospitalisation rates and cumulative incidence were estimated by age and WC group. ResultsUsing Australian Government Centrelink data, WC was defined as parent/s receiving a means-tested welfare payment (low-income, unemployment, disability or caring) when children were aged 11-15, or children receiving payment at ages 16-20. IWC was WC occurring in both parent and child generations. IWC affected 34.9% of children, who had the highest hospitalisation rate (133.5 per 1,000 person-years) compared to no WC (46.1 per 1,000 person-years), parent WC  (75.0 per 1,000 person-years), and child only WC (87.6 per 1,000 person-years). Of all IWC children, 43.0% experienced at least one hospitalisation between 11-20 years, frequently related to injury, mental health, and pregnancy. Conclusions and RelevanceChildren experiencing IWC represent a third of the population aged 11-20. Compared to children with parent-only WC, IWC children had 78% higher hospitalisation rates from age 11 to 20, accounting for over half of all hospitalisations in this age group. Frequent IWC hospitalisation causes were injuries, mental health, and pregnancy.

Evaluation of an Emergency Department Influenza Vaccination Program: Uptake Factors and Opportunities

Introduction: Influenza vaccines are commonly provided through community health events and primary care appointments. However, acute unscheduled healthcare visits such as emergency department (ED) visits are increasingly viewed as important vaccination opportunities. Emergency departments may be well-positioned to complement broader public health efforts with integrated vaccination programs. Methods: We studied an ED-based influenza vaccination initiative in an urban hospital and examined patient-level factors associated with screening and vaccination uptake. Our analyses included patient visits to the ED from October 1, 2019-April 1, 2020. Results: The influenza screening and vaccination program proved feasible. Of the 20,878 ED visits that occurred within the study period, 3,565 (17.1%) included a screening for influenza vaccine eligibility; a small proportion (11.5%) of the patients seen had multiple screenings. Among the patients screened eligible for the vaccine, 916 ultimately received an influenza vaccination while in the ED (43.7% of eligible patients). There was significant variability in the characteristics of patients who were and were not screened and vaccinated. Age, gender, race, preferred language, and receipt of a flu vaccine in prior years were associated with screening and/or receiving a vaccine in the ED. Conclusion: Vaccination programs in the ED can boost community vaccination rates and play a role in both preventing and treating current and future vaccine-preventable public health crises, although efforts must be made to deliver services equitably.

POS0177 HETEROGENOUS CARTILAGE DAMAGE SEEN ON MRI AMONG KNEES WITH KELLGREN-LAWRENCE 2 & 3 OSTEOARTHRITIS: WHAT ARE THE IMPLICATIONS FOR CLINICAL TRIALS?

Background

The most recent update of the Global Burden of Disease figures (GBD 2013) estimated that 242 million people were living in the world with symptomatic and activity-limiting OA of the hip and/or knee. Many potential disease-modifying osteoarthritis drugs (DMOADs) have been investigated, but to date no DMOADs that slow or stop disease progression have been approved by the Food and Drug Administration (FDA) or the European Medicines Agency (EMA). A potential reason for the lack of demonstrated efficacy may be reliance on radiographs for defining structural inclusion and exclusion criteria for clinical trials, such as use of joint space width and Kellgren-Lawrence (KL) grade as surrogates for cartilage damage.

Objectives

To estimate the distribution of cartilage damage seen on knee MRI in a sample of knees with radiographic KL 2 and 3 OA that would potentially qualify for a DMOAD trial.

Methods

We selected knees from the Osteoarthritis Initiative (OAI), a longitudinal cohort study of knees with or at risk of developing symptomatic radiographic OA, that met common structural inclusion criteria for DMOAD trial enrollment at OAI baseline: knees with radiographs centrally graded as KL 2 or 3 and medial minimum joint space width (mJSW) ≥ 1.5mm. A musculoskeletal radiologist with 10 years of experience in semi-quantitative MRI assessment scored knee cartilage damage in the medial and lateral tibiofemoral and patellofemoral compartments using WORMS (Whole-Organ Magnetic Resonance Imaging Score). Coronal intermediate weighted (IW) TSE and sagittal fat-suppressed IW TSE sequences on 3T MRI were used. The WORMS cartilage scores, which are based on both the extent and depth of cartilage damage, were collapsed into 4 categories: no cartilage damage (WORMS 0 and 1), focal partial or full-thickness (PT/FT) cartilage damage (WORMS 2 and 2.5), diffuse partial thickness (PT) cartilage damage (WORMS 3 and 4), and diffuse full-thickness (FT) cartilage damage (WORMS 5 and 6). We estimated the prevalence of each category of cartilage damage in KL2 and KL3 knees; 95% confidence intervals (CI) accounted for clustering at the participant-level since some participants contributed two knees to the analysis.

Results

We identified 2,372 participants contributing 3,446 knees with radiographic OA (KL 2 and 3) and medial mJSW ≥ 1.5mm. There were 2,318 KL2 knees and 1,128 KL3 knees. The distribution of cartilage damage in each compartment by KL grade is presented in Table 1. We found no cartilage damage in any compartments in 9.8% (95%CI: 8.5, 11.1) of KL2 knees and 2.0% (95%CI: 1.1, 2.9) of KL3 knees. Cartilage damage was absent in the medial tibiofemoral compartment in 52.4% (95%CI: 50.1, 54.6) of KL2 knees, and 14.4% (95%CI: 12.2, 16.6) of KL3 knees, versus 61% (95%CI: 58.8, 63.2) of KL2 knees and 53.6% (95%CI: 50.4, 56.7) of KL3 knees in the lateral compartment. When medial and lateral compartments were combined, cartilage damage was absent in 34.8% (95%CI: 32.7, 36.9) of the KL2 knees, and 4.3% (95%CI: 3.0, 5.5) of the KL3 knees. Diffuse FT cartilage lesions in the medial compartment were found in 6.1% (95%CI: 5.0, 7.1) of KL2 knees and 42.5% (95%CI: 39.4, 45.6) of KL3 knees.

Conclusion

MRI screening prior to clinical trial enrollment may identify a substantial percentage of knees with normal cartilage, as well as knees with diffuse FT cartilage lesions that may not be responsive to DMOADs, depending on the mode of action of a given pharmacological compound.

Disclosure of Interests

Mohamed Jarraya: None declared, Frank Roemer Shareholder of: Boston Imaging Core Lab, Consultant of: California Institute of Biomedical Research, Erin Ashbeck: None declared, John Lynch: None declared, C. Kent Kwoh Consultant of: Novartis, Regeneron, LG Chem, Kolon Tissue Gene, Avalor, Grant/research support from: Pfizer, Lilly, Cumberland, Ali Guermazi Shareholder of: Stock options in BICL, Consultant of: Pfizer, TissueGene, MerckSerono, Regeneron, Novartis, AstraZeneca

Isolation, marginalisation and disempowerment - understanding how interactions with health providers can influence smoking cessation in pregnancy

Background

Maternal smoking during pregnancy can lead to serious adverse health outcomes for both women and their infants. While smoking in pregnancy has declined over time, it remains consistently higher in women with lower socioeconomic circumstances. Furthermore, fewer women in this group will successfully quit during pregnancy.

Aim

This study explores the barriers to smoking cessation experienced by socially disadvantaged pregnant women and investigates how interactions with health providers can influence their smoking cessation journey.

Methods

Women (either pregnant or birthed in the previous 10 years, who smoked or quit smoking in pregnancy) were recruited from a metropolitan public hospital antenatal clinic in South Australia and community organisations in surrounding suburbs. Seventeen women participated in qualitative semi-structured small focus groups or interviews. The focus groups and interviews were recorded, transcribed and thematically analysed.

Findings

Four interconnected themes were identified: 1) smoking embedded in women’s challenging lives and pregnancies, 2) cyclic isolation and marginalisation, 3) feeling disempowered, and 4) autonomy and self-determination. Themes 3 and 4 are characterised as being two sides of a single coin in that they coexist simultaneously and are inseparable. A key finding is a strong unanimous desire for smoking cessation in pregnancy but women felt they did not have the necessary support from health providers or confidence and self-efficacy to be successful.

Conclusion

Women would like improvements to antenatal care that increase health practitioners’ understanding of the social and contextual healthcare barriers faced by women who smoke in pregnancy. They seek improved interventions from health providers to make informed choices about smoking cessation and would like women-centred care. Women feel that with greater support, more options for cessation strategies and consistency and encouragement from health providers they could be more successful at antenatal smoking cessation. If such changes were made, then South Australian practice could align more with best practice international guidelines for addressing smoking cessation in pregnancy, and potentially improve outcomes for women and their children.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-022-04720-0.

Prostate Cancer Treatment with Pencil Beam Proton Therapy Using Rectal Spacers sans Endorectal Balloons

Purpose

Proton beam radiotherapy (PBT) has been used for the definitive treatment of localized prostate cancer with low rates of high-grade toxicity and excellent patient-reported quality-of-life metrics. Technological advances such as pencil beam scanning (PBS), Monte Carlo dose calculations, and polyethylene glycol gel rectal spacers have optimized prostate proton therapy. Here, we report the early clinical outcomes of patients treated for localized prostate cancer using modern PBS–PBT with hydrogel rectal spacing and fiducial tracking without the use of endorectal balloons.

Materials and Methods

This is a single institutional review of consecutive patients treated with histologically confirmed localized prostate cancer. Prior to treatment, all patients underwent placement of fiducials into the prostate and insertion of a hydrogel rectal spacer. Patients were typically given a prescription dose of 7920 cGy at 180 cGy per fraction using a Monte Carlo dose calculation algorithm. Acute and late toxicity were evaluated using the Common Terminology Criteria for Adverse Events (CTCAE), version 5. Biochemical failure was defined using the Phoenix definition.

Results

From July 2018 to April 2020, 33 patients were treated (median age, 75 years). No severe acute toxicities were observed. The most common acute toxicity was urinary frequency. With a median follow-up of 18 months, there were no high-grade genitourinary late toxicities; however, one grade 3 gastrointestinal toxicity was observed. Late erectile dysfunction was common. One treatment failure was observed at 21 months in a patient treated for high-risk prostate cancer.

Conclusion

Early clinical outcomes of patients treated with PBS–PBT using Monte Carlo–based planning, fiducial placement, and rectal spacers sans endorectal balloons demonstrate minimal treatment-related toxicity with good oncologic outcomes. Rectal spacer stabilization without the use of endorectal balloons is feasible for the use of PBS–PBT.

Improving hand hygiene practices in two regional hospitals in Kenya using a continuous quality improvement (CQI) approach

Background

Hand hygiene (HH) is central in prevention of health care-associated infections. In low resource settings, models to improve HH compliance are needed. We implemented a continuous quality improvement (CQI) program targeting HH in two hospitals in Kenya.

Objective

To determine the impact of the HH CQI program and identify factors associated with HH compliance between 2018 and 2019.

Methods

A CQI project targeting the improvement of hand hygiene was implemented, including training and mentorship. Data were collected monthly between April 2018 and December 2019 in Thika and Kitale Hospitals. Healthcare workers trained on Infection Prevention and Control (IPC) observed and recorded HH opportunities and subsequent compliance among staff, including nurses, clinicians, and auxiliary staff, using the World Health Organization’s “My Five Moments for Hand Hygiene” tool. Covariates were explored using mixed-effects logistic regression with random department-level intercepts.

Results

Hand hygiene compliance improved from 27% at baseline to 44% after 21 months. Indication/moment for HH was significantly associated with compliance. Adjusting for site, professional category and department, compliance was higher after a moment of body fluid exposure (aOR 1.43, 95% CI 1.17–1.74, p value < 0.001) and lower before an aseptic procedure (aOR 0.12, 95% CI 0.08–0.17, p value < 0.001) compared to after patient contact. Wearing of gloves often replaced proper HH in surgical departments, which although not significant, had lower compliance compared to departments for internal medicine (aOR 0.93, 95% CI 0.85–1.02). Adjusted HH compliance from all quarters improved from baseline, but comparing each quarter to the previous quarter, the improvement fluctuated over time.

Conclusion

Training and mentorship on the importance of HH for all moments is needed to improve overall HH compliance. CQI with regular monitoring and feedback of HH performance can be an effective approach in improving HH compliance in public hospitals in Kenya.

Observed Versus Expected COVID-19 Infections Among National Football League Players During the 2020 Season

BACKGROUND

The National Football League (NFL) and National Football League Players Association implemented a set of strict protocols for the 2020 season with the intent to mitigate COVID-19 risk among players and staff. In that timeframe, the league's 32 teams completed 256 regular season games and several thousand meetings and practices. In parallel, community cases of COVID-19 were highly prevalent. We assess the risk of holding a 2020 NFL season by comparing community and player COVID-19 infections.

METHODS

We used county-level COVID-19 test data from each team to establish baseline distributions of infection rates expected to occur in a population similar in age and sex to NFL players. We used a binomial distribution to simulate expected infections in each community cohort and compared these findings with observed COVID-19 infections in players.

RESULTS

Over a 5-month period (1 August 2020 to 2 January 2021), positive NFL player infections (n = 256) were 55.7% lower than expected when compared with simulations from NFL community cohorts. For 30 of 32 teams (94%), observed counts fell at or below expectation, including 28 teams (88%) for which rates were lower. Two teams fell above baseline expectation.

CONCLUSIONS

The NFL/NFLPA protocols that governed team facilities, travel, gameday, and activities outside of the workplace were associated with lower infection rates among NFL players compared with the surrounding community. The NFL's 2020-2021 season are consistent with the hypothesis that robust testing and behavioral protocols support a safe return to sport and work.

Methane and the Paris Agreement temperature goals

Meeting the Paris Agreement temperature goal necessitates limiting methane (CH4)-induced warming, in addition to achieving net-zero or (net-negative) carbon dioxide (CO2) emissions. In our model, for the median 1.5°C scenario between 2020 and 2050, CH4 mitigation lowers temperatures by 0.1°C; CO2 increases it by 0.2°C. CO2 emissions continue increasing global mean temperature until net-zero emissions are reached, with potential for lowering temperatures with net-negative emissions. By contrast, reducing CH4 emissions starts to reverse CH4-induced warming within a few decades. These differences are hidden when framing climate mitigation using annual 'CO2-equivalent' emissions, including targets based on aggregated annual emission rates. We show how the different warming responses to CO2 and CH4 emissions can be accurately aggregated to estimate warming by using 'warming-equivalent emissions', which provide a transparent and convenient method to inform policies and measures for mitigation, or demonstrate progress towards a temperature goal. The method presented (GWP*) uses well-established climate science concepts to relate GWP100 to temperature, as a simple proxy for a climate model. The use of warming-equivalent emissions for nationally determined contributions and long-term strategies would enhance the transparency of stocktakes of progress towards a long-term temperature goal, compared to the use of standard equivalence methods. This article is part of a discussion meeting issue 'Rising methane: is warming feeding warming? (part 2)'.

Magnetic Resonance Imaging-Assessed Subchondral Cysts and Incident Knee Pain and Knee Osteoarthritis: Data From the Multicenter Osteoarthritis Study

OBJECTIVE

To examine whether knee subchondral cysts, measured on magnetic resonance imaging (MRI), are associated with incident knee osteoarthritis (OA) outcomes.

METHODS

We used longitudinal data from the Multicenter Osteoarthritis Study, a community-based cohort of subjects with risk factors for knee OA. Participants without a history of knee surgery and/or inflammatory arthritis (i.e., rheumatoid arthritis and gout) were followed up for 84 months for the following incident outcomes: 1) radiographic knee OA (Kellgren/Lawrence grade ≥2), 2) symptomatic radiographic knee OA (radiographic knee OA and frequent knee pain), and 3) frequent knee pain (with or without radiographic knee OA). In a subset of participants, subchondral cysts were scored on baseline MRIs of 1 knee. Multiple logistic regression, with adjustment for participant characteristics and other baseline knee MRI findings, was used to assess whether subchondral cysts were predictive of incident outcomes.

RESULTS

Among the participants with knees eligible for analyses of outcomes over 84 months, incident radiographic knee OA occurred in 22.8% of knees with no baseline radiographic knee OA, symptomatic radiographic knee OA occurred in 17.0% of knees with no baseline symptomatic radiographic knee OA, and frequent knee pain (with or without radiographic knee OA) occurred in 28.8% of knees with no baseline radiographic knee OA and 43.7% of knees with baseline radiographic knee OA. With adjustment for age, sex, and body mass index, the presence of subchondral cysts was not associated with incident radiographic knee OA but was associated with increased odds of incident symptomatic radiographic knee OA (odds ratio 1.92 [95% confidence interval 1.16-3.19]) and increased odds of incident frequent knee pain in those who had radiographic knee OA at baseline (odds ratio 2.11 [95% confidence interval 0.87-5.12]). Stronger and significant associations were observed for outcomes based on consistent reports of frequent knee pain within ~1 month of the study visit.

CONCLUSION

Subchondral cysts are likely to be a secondary phenomenon, rather than a primary trigger, of radiographic knee OA, and may predict symptoms in knees with existing disease.

Short- and long-term warming effects of methane may affect the cost-effectiveness of mitigation policies and benefits of low-meat diets

Methane’s short atmospheric life has important implications for the design of global climate change mitigation policies in agriculture. Three different agricultural economic models are used to explore how short- and long-term warming effects of methane can affect the cost-effectiveness of mitigation policies and dietary transitions. Results show that the choice of a particular metric for methane’s warming potential is key to determine optimal mitigation options, with metrics based on shorter-term impacts leading to greater overall emission reduction. Also, the promotion of low-meat diets is more effective at reducing greenhouse gas emissions compared to carbon pricing when mitigation policies are based on metrics that reflect methane’s long-term behaviour. A combination of stringent mitigation measures and dietary changes could achieve substantial emission reduction levels, helping reverse the contribution of agriculture to global warming.

The politics of ageing: how to get policymakers to support lifecourse policies

Background

Given that there is not much evidence that ageing imperils the finance and provision of health care, why do so many policymakers act like it does?

Methods

We break conventional wisdom down into myths and realities, identifying the evidence against them.

Results

A first myth is that ageing produces unsustainable health care costs, which in turn, creates intergenerational conflict over public policy. A second myth is that older people behave as a single group, always pursuing policies that benefit themselves. The final myth is that decisions about policy are made by politicians who pander to that elderly block. The first reality is that most of the problems ascribed to inequality between generations (intergenerational equity) are actually problems of inequality within society as a whole that span across age groups (intragenerational equity). The second reality is that policies that address these broader inequalities are built on the life-course perspective, which focuses on identifying the policies which can make people happier and healthier at all ages by drawing on the context and circumstances under which aging occurs. The third reality is that it is possible to construct coalitions of politicians and interests that can develop and support sophisticated life-course policies that lessen the burdens of ageing and health on everybody.

Conclusions

Intergenerational inequality is not, and need not be, a significant problem for rich countries. It is substantially a product of current and past intragenerational inequality, and in fact inequality between generations often goes with inequality within generations. Intergenerational conflict is a distraction from policies that promote greater equality within and between generations, and talk of an ageing crisis is frequently just another version of longstanding arguments against public social investment from cradle to grave.