Emily Griffiths

Coinfection by multiple parasites affects human health, parasite dynamics and the efficacy of infectious disease prevention and treatment. The capacity for different parasites (viruses, bacteria, fungi, protozoa, helminths) to interact is poorly understood. Interspecific interactions between coinfecting parasites could occur in many ways, either directly or indirectly with the host's immune system or bodily resources. We aimed to summarise connected resources, parasites and immune system components in coinfected humans using a network approach. The published literature contains thousands of records of coinfections in humans, associated immune responses, as well as parasite resource requirements. We used over 250 publications on human coinfection from 2009 to build an evidence-based parasite-immune system-resource network. We recorded the identity of coinfecting parasites, immune system components, host resources and the reported relationships between them. The network represents the potential for parasites to interact based on observation and theory found in recent coinfection literature. Results show the great taxonomic variety of coinfecting parasites, with particular involvement of viruses. Some parasites were reported in more coinfections, most notably HIV, suggesting that recent disease invasion and induced immunodeficiency may facilitate many parasite interactions. The network can also be used to generate numerous hypotheses for modelling work and suggestions for future observational and experimental research. The use of networks and other research tools to understand parasite interactions within coinfected hosts will help predict the potential for and consequences of disease invasions, as well as improve infectious disease interventions.

Many people have multiple infections at the same time, but the combined contribution of those infections to disease-related mortality is unknown. Registered causes
of death offer a unique opportunity to study associations between multiple infections. We analysed over 900,000 death certificates that reported infectious causes of death. We tested whether reports of multiple infections (ie, co-infections) differed across individuals' age or sex. We also tested whether each pair of infections were reported together more or less often than expected by chance, and whether this co-reporting was associated with the number of biological characteristics they had in common.

Simultaneous infection by multiple parasite species (viruses, bacteria, helminths, protozoa or fungi) is commonplace. Most reports show coinfected humans to have worse health than those with single infections. However, we have little understanding of how coinfecting parasites interact within human hosts. We used data from over 300 published studies to construct a network that offers the first broad indications of how groups of coinfecting parasites tend to interact. The network had three levels comprising parasites, the resources they consume, and the immune responses they elicit, connected by potential, observed, and experimentally proven links. Pairs of parasite species had most potential to interact indirectly through shared resources, rather than through immune responses or other parasites. Also, the network comprised 10 tightly knit groups, eight of which were associated with particular body parts, and seven of which were dominated by parasite-resource links. Coinfection in humans is therefore structured by physical location within the body, with bottom-up, resource-mediated processes most often influencing how, where, and which coinfecting parasites interact. The many indirect interactions show how treating an infection could affect other infections in coinfected patients, but the compartmentalised structure of the network will limit how far these indirect effects are likely to spread.

Many fundamental patterns of coinfection (multi-species infections) are undescribed, including the relative frequency of coinfection by various pathogens, differences between single-species infections and coinfection, and the burden of coinfection on human health. We aimed to address the paucity of general knowledge on coinfection by systematically collating and analysing data from recent publications to understand the types of coinfection and their effects.

From an electronic search to find all publications from 2009 on coinfection and its synonyms in humans we recorded data on i) coinfecting pathogens and their effect on ii) host health and iii) intensity of infection.

The most commonly reported coinfections differ from infections causing highest global mortality, with a notable lack of serious childhood infections in reported coinfections. We found that coinfection is generally reported to worsen human health (76% publications) and exacerbate infections (57% publications). Reported coinfections included all kinds of pathogens, but were most likely to contain bacteria.

These results suggest differences between coinfected patients and those with single infections, with coinfection having serious health effects. There is a pressing need to quantify the tendency towards negative effects and to evaluate any sampling biases in the coverage of coinfection research.

The number of psychiatric hospital beds in England has declined since the 1950s. Since the early 2000s mental health staff increasingly work in community treatment teams. We analysed recent trends in hospital and community treatment in England for eight mental health diagnoses.

We obtained data from the UK Government Health and Social Care Information Centre covering the period 1998 to 2012. We analysed hospital admissions and length of stay for each diagnosis each year using linear regression. We studied associations among admissions, community treatment, and hospital bed availability each year using structural equation modeling.

The number of mental health beds fell 39%, from 37000 in 1998 to 22300 in 2012. Hospital admissions for five diagnoses declined significantly (depression, bipolar disorder, schizophrenia, dementia and Obsessive Compulsive Disorder, p<0.01 or p<0.001). The strongest decline for depression involved 1000 fewer admissions each year. Admissions for three disorders increased significantly (Post Traumatic Stress Disorder, eating disorders and alcohol-related disorders, p<0.01 or p<0.001). Alcohol-related admissions increased most strongly, by more than 1700 a year, and were significantly associated with increasing liver fibrosis and cirrhosis admissions (Pearson’s r=0.89, p<0.001) across the NHS, and the affordability of alcohol (Pearson’s r =0.76, p<0.01).

The median length of stay declined significantly for four diagnoses (p<0.001); the other four diagnoses did not change significantly. Depression had the steepest decline of almost one less day in hospital per admission per year.

Almost 300 more patients were sectioned under the Mental Health Act each year. Community activity had relatively little effect on admissions, and its direct effect was not significantly different from zero. Years with more psychiatric beds had more admissions.

Mental health bed numbers have declined significantly in England. Annual admissions and lengths of stay declined for a range of severe mental disorders including schizophrenia, bipolar disorder, and depression. The fall in available beds can account for much of the decline in admissions. National reports of crisis team activity are not associated with declines in hospital admissions. There may be significant needs, especially of depressive patients, not being met by secondary community services, such as 24-hour observation and care. This calls for policy review and further epidemiological study of morbidity, mortality and health needs associated with mental disorder in the community.