Risk Factors and Impact of Patterns of Co-Occurring Comorbidities in People Living With HIV

Davide De Francesco; Jonathan Underwood; Emmanouil Bagkeris; Jane Anderson; Ian Williams; Jaime H. Vera; Frank A. Post; Marta Boffito; Margaret Johnson; Patrick W.G. Mallon; Alan Winston; Caroline A. Sabin; on behalf of the Pharmacokinetic and Clinical Observations in PeoPle Over fiftY (POPPY) study


AIDS. 2019;33(12):1871-1880. 

In This Article

Abstract and Introduction


Aims: To assess associations of comorbidity patterns observed in people living with HIV (PLWH) with risk factors and health outcomes.

Methods: Common patters of comorbidities in PLWH participating in the Pharmacokinetic and Clinical Observations in People Over Fifty study were determined using principal component analysis and a severity score for each pattern was derived. Associations between each pattern's severity score and risk factors were assessed using median regression. The independent associations of patterns' severity scores with self-reported physical and mental health (SF-36 summary scores) were assessed using linear regression, with functional impairment (Lawton IADL < 8) and hospitalization in last year using logistic regression and with number of general practitioner visits using Poisson regression.

Results: A total of 1073 PLWH were analysed: 85.2% male, median (interquartile range) age 52 (47–59) years, 98% on therapy. Duration of HIV was associated with higher severity in 4/6 of patterns: cardiovascular diseases, mental health problems, metabolic disorders and chest/other infections (all P ≤ 0.001). Prior AIDS was associated with higher severity scores for the same patterns and for the pattern of cancers (P < 0.001). The pattern of cardiovascular diseases was associated with poorer physical health (P = 0.02), higher risk of functional impairment (P = 0.02) and hospitalization (P < 0.001) and with higher number of general practitioner visits (P < 0.001). Severity of mental health (all P < 0.001) and of chest/other infections patterns negatively affected all the five health outcomes.

Conclusion: Common patterns of comorbidities seen in PLWH appear to have different risk factors and to differently affect health outcomes. These findings may assist the development of targeted intervention to prevent, treat and manage the increasingly prevalent multimorbidity in PLWH.


The widespread use of combination antiretroviral treatment has led to an increase in life expectancy of people living with HIV (PLWH) in many regions of the world,[1] resulting in major demographic changes that are expected to continue. For example, studies have estimated that up to 73% of PLWH in the Netherlands will be aged 50 years or older in 2030,[2] with similar proportions in Italy and the USA.[3] As a result, PLWH are experiencing an increasing burden of comorbidities[4] that is also expected to further increase; the same studies have estimated that the proportions of PLWH with at least one comorbidity would reach 84% in the Netherlands by 2030 and 89% in Italy and the USA by 2035.

We have reported a tendency for comorbidities to co-occur in the same HIV-positive individual at prevelances that are higher than those expected by chance alone and to occur in specific patterns.[5] Although knowledge of frequent patterns of comorbidities could inform decisions for prevention strategies, a better understanding of the mechanistic basis underlying these patterns and the burden they pose on PLWH and on the healthcare system is required.

Both in the general population and in populations of PLWH, most of the research published to date has focused on determinants of multimorbidity rather than of specific patterns. Among PLWH, studies have reported associations of multimorbidity with age and obesity,[6] smoking[7] and duration of HIV infection.[8,9] However, considering the diverse nature of conditions and comorbidities that could be encompassed into a definition of multimorbidity, these findings are unlikely to provide generalizable evidence of the causes underlying specific patterns. The two previous studies that focused on data-driven patterns of comorbidities in PLWH only investigated associations with one selected factor: Goulet et al.[10] found hepatits C virus coinfection to be associated with patterns of mental health problems and alcohol-related complications, while Kim et al.[11] reported an association of obesity with metabolic and behavioural problems. Little is known about the impact of HIV-related factors and other non-HIV-related modifiable and nonmodifiable factors on specific patterns of comorbidities commonly seen in PLWH.

Similarly, while there is some evidence of the impact of multimorbidity on health outcomes such as quality of life of PLWH[12] and healthcare costs,[13] there are no data on the impact of specific patterns of comorbidities to reveal those with the greatest burden for both patients and the healthcare system. Different patterns of comorbidities can be expected to differentially affect quality of life and use of healthcare resources. Moreover, the effect of a certain pattern can be greater (or lower) than the sum of the independent effects of all comorbidities within the pattern.

The aims of this study were to investigate HIV-related and non-HIV-related risk factors for common patterns of comorbidities observed in PLWH seen for care in the UK and Ireland, and to evaluate associations of each pattern with self-reported health outcomes such as quality of life, functional impairment and healthcare resource use.