Chris Bonell

Process evaluation of complex interventions: Medical Research Council guidance

Process evaluation is an essential part of designing and testing complex interventions. New MRC guidance provides a framework for conducting and reporting process evaluation studies

Process evaluation in randomised controlled trials of complex interventions

Most randomised controlled trials focus on outcomes, not on the processes involved in implementing an intervention. Using an example from school based health promotion, this paper argues that including a process evaluation would improve the science of many randomised controlled trials

Global burden of diseases, injuries, and risk factors for young people's health during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.

BACKGROUND Young peoples health has emerged as a neglected yet pressing issue in global development. Changing patterns of young peoples health have the potential to undermine future population health as well as global economic development unless timely and effective strategies are put into place. We report the past, present, and anticipated burden of disease in young people aged 10-24 years from 1990 to 2013 using data on mortality, disability, injuries, and health risk factors. METHODS The Global Burden of Disease Study 2013 (GBD 2013) includes annual assessments for 188 countries from 1990 to 2013, covering 306 diseases and injuries, 1233 sequelae, and 79 risk factors. We used the comparative risk assessment approach to assess how much of the burden of disease reported in a given year can be attributed to past exposure to a risk. We estimated attributable burden by comparing observed health outcomes with those that would have been observed if an alternative or counterfactual level of exposure had occurred in the past. We applied the same method to previous years to allow comparisons from 1990 to 2013. We cross-tabulated the quantiles of disability-adjusted life-years (DALYs) by quintiles of DALYs annual increase from 1990 to 2013 to show rates of DALYs increase by burden. We used the GBD 2013 hierarchy of causes that organises 306 diseases and injuries into four levels of classification. Level one distinguishes three broad categories: first, communicable, maternal, neonatal, and nutritional disorders; second, non-communicable diseases; and third, injuries. Level two has 21 mutually exclusive and collectively exhaustive categories, level three has 163 categories, and level four has 254 categories. FINDINGS The leading causes of death in 2013 for young people aged 10-14 years were HIV/AIDS, road injuries, and drowning (25·2%), whereas transport injuries were the leading cause of death for ages 15-19 years (14·2%) and 20-24 years (15·6%). Maternal disorders were the highest cause of death for young women aged 20-24 years (17·1%) and the fourth highest for girls aged 15-19 years (11·5%) in 2013. Unsafe sex as a risk factor for DALYs increased from the 13th rank to the second for both sexes aged 15-19 years from 1990 to 2013. Alcohol misuse was the highest risk factor for DALYs (7·0% overall, 10·5% for males, and 2·7% for females) for young people aged 20-24 years, whereas drug use accounted for 2·7% (3·3% for males and 2·0% for females). The contribution of risk factors varied between and within countries. For example, for ages 20-24 years, drug use was highest in Qatar and accounted for 4·9% of DALYs, followed by 4·8% in the United Arab Emirates, whereas alcohol use was highest in Russia and accounted for 21·4%, followed by 21·0% in Belarus. Alcohol accounted for 9·0% (ranging from 4·2% in Hong Kong to 11·3% in Shandong) in China and 11·6% (ranging from 10·1% in Aguascalientes to 14·9% in Chihuahua) of DALYs in Mexico for young people aged 20-24 years. Alcohol and drug use in those aged 10-24 years had an annual rate of change of >1·0% from 1990 to 2013 and accounted for more than 3·1% of DALYs. INTERPRETATION Our findings call for increased efforts to improve health and reduce the burden of disease and risks for diseases in later life in young people. Moreover, because of the large variations between countries in risks and burden, a global approach to improve health during this important period of life will fail unless the particularities of each country are taken into account. Finally, our results call for a strategy to overcome the financial and technical barriers to adequately capture young peoples health risk factors and their determinants in health information systems. FUNDING Bill & Melinda Gates Foundation.

Systematic review exploring time trends in the association between educational attainment and risk of HIV infection in sub-Saharan Africa.

Objective:To assess the evidence that the association between educational attainment and risk of HIV infection is changing over time in sub-Saharan Africa. Design and methods:Systematic review of published peer-reviewed articles. Articles were identified that reported original data comparing individually measured educational attainment and HIV status among at least 300 individuals representative of the general population of countries or regions of sub-Saharan Africa. Statistical analyses were required to adjust for potential confounders but not over-adjust for variables on the causal pathway. Results:Approximately 4000 abstracts and 1200 full papers were reviewed. Thirty-six articles were included in the study, containing data on 72 discrete populations from 11 countries between 1987 and 2003, representing over 200 000 individuals. Studies on data collected prior to 1996 generally found either no association or the highest risk of HIV infection among the most educated. Studies conducted from 1996 onwards were more likely to find a lower risk of HIV infection among the most educated. Where data over time were available, HIV prevalence fell more consistently among highly educated groups than among less educated groups, in whom HIV prevalence sometimes rose while overall population prevalence was falling. In several populations, associations suggesting greater HIV risk in the more educated at earlier time points were replaced by weaker associations later. Discussion:HIV infections appear to be shifting towards higher prevalence among the least educated in sub-Saharan Africa, reversing previous patterns. Policy responses that ensure HIV-prevention measures reach all strata of society and increase education levels are urgently needed.

Alternatives to randomisation in the evaluation of public health interventions: design challenges and solutions

Background There has been a recent increase in interest in alternatives to randomisation in the evaluation of public health interventions. We aim to describe specific scenarios in which randomised trials may not be possible and describe, exemplify and assess alternative strategies. Methods Non-systematic exploratory review. Results In many scenarios barriers are surmountable so that randomised trials (including stepped-wedge and crossover trials) are possible. It is possible to rank alternative designs but context will also determine which choices are preferable. Evidence from non-randomised designs is more convincing when confounders are well-understood, measured and controlled; there is evidence for causal pathways linking intervention and outcomes and/or against other pathways explaining outcomes; and effect sizes are large. Conclusion Non-randomised trials might provide adequate evidence to inform decisions when interventions are demonstrably feasible and acceptable, and where evidence suggests there is little potential for harm, but caution that such designs may not provide adequate evidence when intervention feasibility or acceptability is doubtful, and where existing evidence suggests benefits may be marginal and/or harms possible.

The association between school attendance, HIV infection and sexual behaviour among young people in rural South Africa

Objectives: To investigate whether the prevalence of HIV infection among young people, and sexual behaviours associated with increased HIV risk, are differentially distributed between students and those not attending school or college. Design: A random population sample of unmarried young people (916 males, 1003 females) aged 14–25 years from rural South Africa in 2001. Methods: Data on school attendance and HIV risk characteristics came from structured face-to-face interviews. HIV serostatus was assessed by oral fluid ELISA. Logistic regression models specified HIV serostatus and high-risk behaviours as outcome variables. The primary exposure was school attendance. Models were adjusted for potential confounders. Results: HIV knowledge, communication about sex and HIV testing were similarly distributed among students and non-students. The lifetime number of partners was lower for students of both sexes (adjusted odds ratio (aOR) for more than three partners for men 0.67; 95% CI 0.44 to 1.00; aOR for more than two partners for women 0.69; 95% CI 0.46 to 1.04). Among young women, fewer students reported having partners more than three years older than themselves (aOR 0.58; 95% CI 0.37 to 0.92), having sex more than five times with a partner (aOR 0.57; 95% CI 0.37 to 0.87) and unprotected intercourse during the past year (aOR 0.60; 95% CI 0.40 to 0.91). Male students were less likely to be HIV positive than non-students (aOR 0.21; 95% CI 0.06 to 0.71). Conclusions: Attending school was associated with lower-risk sexual behaviours and, among young men, lower HIV prevalence. Secondary school attendance may influence the structure of sexual networks and reduce HIV risk. Maximising school attendance may reduce HIV transmission among young people.

Assessment of generalisability in trials of health interventions: suggested framework and systematic review

Most evaluations of new treatments use highly selected populations, making it difficult to decide whether they would work elsewhere. Systematic evaluation and reporting of applicability is required

Process evaluation in complex public health intervention studies: the need for guidance

Public health interventions aim to improve the health of populations or at-risk subgroups. Problems targeted by such interventions, such as diet and smoking, involve complex multifactorial aetiology. Interventions will often aim to address more than one cause simultaneously, targeting factors at multiple levels (eg, individual, interpersonal, organisational), and comprising several components which interact to affect more than one outcome.1 They will often be delivered in systems which respond in unpredictable ways to the new intervention.2 Recognition is growing that evaluations need to understand this complexity if they are to inform future intervention development, or efforts to apply the same intervention in another setting or population.1 Achieving this will require evaluators to move beyond a ‘does it work?’ focus, towards combining outcomes and process evaluation. There is no such thing as a typical process evaluation, with the term applied to studies which range from a few simple quantitative items on satisfaction, to complex mixed-method studies exploring issues such as the process of implementation, or contextual influences on implementation and outcomes. As recognised within MRC guidance for evaluating complex interventions, process evaluation may be used to ‘assess fidelity and quality of implementation , clarify causal mechanisms and identify contextual factors associated with variation in outcomes’.1 This paper briefly discusses each of these core aims for process evaluation, before describing current Medical Research Council (MRC) Population Health Sciences Research Network (PHSRN) funded work to develop guidance for process evaluations of complex public health interventions. ### Intervention implementation An important role for process evaluations is to examine the quantity and quality of what was actually implemented in practice, and why. This may inform implementation of similar interventions elsewhere, and facilitate interpretation of intervention outcomes. While notions of standardisation are central to implementation assessment, the nature of …

You are what your friends eat: systematic review of social network analyses of young people's eating behaviours and bodyweight

Background This review synthesises evidence regarding associations between young peoples social networks and their eating behaviours/bodyweight, and also explores how these vary according to the setting and sample characteristics. Methods A systematic review of cross-sectional and longitudinal observational studies examining the association between measures of young peoples social networks based on sociometric data and eating behaviours (including calorific intake) and/or bodyweight. Results There is consistent evidence that school friends are significantly similar in terms of their body mass index, and friends with the highest body mass index appear to be most similar. Overweight youth are also less likely to be popular and more likely to be socially isolated at school. Frequency of fast food consumption has also been found to cluster within groups of boys, as have body image concerns, dieting and eating disorders among girls. Conclusion School friendships may be critical in shaping young peoples eating behaviours and bodyweight and/or vice versa, and suggests the potential of social-network-based health promotion interventions in schools. Further longitudinal research is needed to examine the processes via which this clustering occurs, how it varies according to school context, and the effects of non-school networks.

Explaining continued high HIV prevalence in South Africa: socioeconomic factors, HIV incidence and sexual behaviour change among a rural cohort, 2001-2004

Objectives:To estimate HIV incidence and explore evidence for changing sexual behaviour over time among men and women belonging to different socioeconomic groups in rural South Africa. Design and methods:A cohort study conducted between 2001 and 2004; 3881 individuals aged 14–35 years enumerated in eight villages were eligible. At least three household visits were made to contact each eligible respondent at both timepoints. Sexual behaviour data were collected in structured, respondent-focused interviews. HIV serostatus was assessed using an oral fluid enzyme-linked immunosorbent assay at each timepoint. Results:Data on sexual behaviour were available from 1967 individuals at both timepoints. A total of 1286 HIV-negative individuals at baseline contributed to the analysis of incidence. HIV incidence was 2.2/100 person-years among men and 4.9/100 person-years in women, among whom it was highest in the least educated group. Median age at first sex was lower among later birth cohorts. A higher number of previously sexually active individuals reported having multiple partners in the past year in 2004 than 2001. Condom use with non-spousal partners increased from 2001 to 2004. Migrant men more often reported multiple partners. Migrant and more educated individuals of both sexes and women from wealthier households reported higher levels of condom use. Discussion:HIV incidence is high in rural South Africa, particularly among women of low education. Some risky sexual behaviours (early sexual debut, having multiple sexual partners) are becoming more common over time. Condom use is increasing. Existing HIV prevention strategies have only been partly effective in generating population-level behavioural change.