Food-based dietary guidelines have been the basis of public health recommendations for over half a century, but more recently, there has been a trend to classify the health properties of food not by its nutrient composition, but by the degree to which it has been processed. This concept has been supported by many association studies, narrative reviews and the findings from one randomised controlled feeding trial, which demonstrated the sustained effect of ultra-processed diets on increasing both energy intake and body weight. This has led to widespread speculation as to specific features of ultra-processed foods that promote increased energy intakes. Rising interest in the ultra-processed topic has led to proposals to include guidance and restrictions on the consumption of processed foods in national dietary guidelines, with some countries encouraging consumers to avoid highly processed foods completely, and only choose minimally processed foods. However, there remains a lack of consensus on the role of processed foods in human health when faced with the challenges of securing the food supply for a growing global population, that is, healthy, affordable and sustainable. There has also been criticism of the subjective nature of definitions used to differentiate foods by their degree of processing, and there is currently a lack of empirical data to support a clear mechanism by which highly processed foods promote greater energy intakes. Recommendations to avoid all highly processed foods are potentially harmful if they remove affordable sources of nutrients and will be impractical for most when an estimated two-thirds of current energy purchased are from processed or ultra-processed foods. The current review highlights some considerations when interpreting the dietary association studies that link processed food intake to health and offers a critique on some of the mechanisms proposed to explain the link between ultra-processed food and poor health. Recent research suggests a combination of higher energy density and faster meal eating rates are likely to influence meal size and energy intakes from processed foods and offers new perspectives on how to manage this in the future. In going beyond the ultra-processed debate, the aim is to summarise some important considerations when interpreting existing data and identify the important gaps for future research on the role of processed food in health.

The goal of much observational research is to identify risk factors that have a causal effect on health and social outcomes. However, observational data are subject to biases from confounding, selection and measurement, which can result in an underestimate or overestimate of the effect of interest. Various advanced statistical approaches exist that offer certain advantages in terms of addressing these potential biases. However, although these statistical approaches have different underlying statistical assumptions, in practice they cannot always completely remove key sources of bias; therefore, using design-based approaches to improve causal inference is also important. Here it is the design of the study that addresses the problem of potential bias – either by ensuring it is not present (under certain assumptions) or by comparing results across methods with different sources and direction of potential bias. The distinction between statistical and design-based approaches is not an absolute one, but it provides a framework for triangulation – the thoughtful application of multiple approaches (e.g. statistical and design based), each with their own strengths and weaknesses, and in particular sources and directions of bias. It is unlikely that any single method can provide a definite answer to a causal question, but the triangulation of evidence provided by different approaches can provide a stronger basis for causal inference. Triangulation can be considered part of wider efforts to improve the transparency and robustness of scientific research, and the wider scientific infrastructure and system of incentives.

Whether men's and women's reciprocation of their intimate partners’ negative and positive affect during conflictual topic discussions accounted for the association between their trait hostility and perpetration of physical intimate partner violence (IPV) was examined within a dyadic model, using concurrent measurement. The work builds on that of Dr. Tom Dishion regarding hostile and coercive interactions in key relationships on risk outcomes and the importance of moment-by-moment influences in social interactions. Using dynamic development systems theory and a community sample of at-risk men (N = 156) and their female partners, the hypothesis that quicker negative and slower positive affect reactivity would account for physical IPV perpetration beyond trait hostility was tested. Results suggest that, for women, quicker negative affect reactivity partially explains the hostility IPV association, whereas for men, trait hostility of both partners best explained their perpetration of physical IPV. No support was found for positive affect reactivity as a protective relationship process for IPV involvement. Findings are in line with other studies indicating men were less likely to engage in negative reciprocity relative to women. Furthermore, findings highlight how both partners’ individual characteristics, communication patterns, and emotion regulation processes germane to the romantic relationship impact the likelihood of experiencing physical IPV.

New studies of long-term outcomes claim to show that taking antipsychotics on a continuous and indefinite basis is the best approach for people diagnosed with a first episode of psychosis or schizophrenia. A 10-year follow-up of a trial of quetiapine maintenance, for example, found a higher proportion of people with a poor composite outcome in the group initially randomised to placebo. However, most people classified as showing poor outcome were rated as having a mild score on a single psychotic symptom; there were no differences in overall symptoms, positive or negative symptoms or level of functioning. Moreover, 16% of participants did not have a follow-up interview and data from the end of the original trial were used instead. A study using a Finnish database suggested that mortality and readmission were higher in people who did not start long-term antipsychotic treatment or who discontinued it as compared with long-term continuous users. However, the analysis did not control for important confounders and is likely to reflect the fact that people who do not comply with treatment are at higher risk of death due to underlying health risks and behaviours. The analysis showed a slightly higher risk of readmission among non-users of antipsychotics compared with long-term users and a more substantial increased risk among people who discontinued treatment. However, follow-up ceased at the first readmission and therefore eventual, long-term outcome was not assessed. Speed of reduction and whether it was done with or without clinical support were also not distinguished.