Addressing the social significance of HIV

George Shirreff

george.shirreff07@ic.ac.uk. Department of Infectious Disease Epidemiology, Imperial College, London

Addressing the social significance of HIV is essential in understanding and preventing its spread and alleviating its impact on the population.

HIV infections are often most prevalent among hard-to-reach populations such as intravenous drug users and commercial sex workers. This creates challenges for identifying and treating individual infections, and it is also difficult to estimate the prevalence of the infection and the effectiveness of interventions. Many with HIV infection experience considerable stigma and this is a barrier both to human well-being and to control of the epidemic. Social reactions to HIV are also important in assessing the impact of changes to behaviour such as condom use or other sexual risk factors. For example, how does condom use change following a test, either with a positive or negative result.

In most cases, the spread of HIV is highly dependent on sex and therefore sexual networks. Network science is an emerging field and is essential to epidemiology of HIV infection. The spread of HIV through sexual networks is dependent on many aspects of the network such as number of partners, heterogeneity of sexual behaviour and concurrency of multiple partnerships. The spread of other sexually transmitted infections is also important as a risk factor for HIV.

In countries where HIV prevalence is extremely high, as in sub-Saharan Africa, the infection devastates the working age population. Large quantities of orphans and a shortage of teachers are among the more direct effects of the premature deaths caused by AIDS.

Development of a new set of criteria to assess the quality of diverse studies as part of a systematic review

Reema Sirriyeh

R.H.Sirriyeh07@leeds.ac.uk. Institute of Psychological Sciences, University of Leeds

Whilst attempting to conduct a systematic review of research relating to the impact of medical error on the health professional, the lack of a suitable tool to assess study quality where study designs are diverse became apparent. A number of existing scales were examined which led to the identification of a set of 17 criteria (Jackson, Cheater & Reid, 2008) that had been developed for diverse studies based on prior attempts (Edwards, Hood, Matthews et al, 2000; Thomas & Harden, 2008; Thomas, Sutcliffe, Harden et al, 2003). However, difficulties arose in their application due to the dichotomous nature of assessments, reliance on reviewer’s judgements and omissions within the criteria. Therefore, the aim of this project was to combine existing strategies with expert opinion to devise a tool to be used in systematic reviews for the assessment of study quality where study designs are diverse.

The original 17 criteria were adapted to form a new set of 16 new criteria. Each item was scored using a scaled response from 0 ‘no mention at all’ – 3 ’complete’, to indicate the extent to which the issue had been covered in each paper. Guidelines for scoring each item were developed to minimise the level of judgement or expertise required from reviewers. The tool was trialled with 3 reviewers and a random sample of the papers from the review (4). Following 2 stages of revisions and refinements to the tool, substantial agreement (Landis & Koch, 1977) was established between 3 reviewers (κ = 62.9%) and the scale was applied to all the 25 papers in the review. Review findings indicated that the quality of evidence was poor in this set of papers, achieving only 47.8% of the total possible maximum score. Despite descriptive components of the papers scoring highly, key weaknesses were identified in aspects of study design such as sample size, and the quality of the measurement tools that were employed.

This project has established a quality assessment tool to establish the quality of evidence where studies are diverse. Such a tool is valuable for much health and social research where variation in study design often raises challenges in determining the strength of evidence in a body of research. Further work to determine the validity and usability of this tool continues.