Impact of late presentation of HIV infection on short-, mid- and long-term mortality and causes of death in a multicenter national cohort: 2004–2013

• Published in: The Journal of infection Vol. 72; no. 5; pp. 587 - 596
• Main Authors: Sobrino-Vegas, Paz, Moreno, Santiago, Rubio, Rafael, Viciana, Pompeyo, Bernardino, José Ignacio, Blanco, José Ramón, Bernal, Enrique, Asensi, Víctor, Pulido, Federico, del Amo, Julia, Hernando, Victoria
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.05.2016
• Subjects: Adult; Aged; Cause of Death; Causes of death; Cohort Studies; Cohort study; Delayed Diagnosis; Factors and trends; Female; HIV; HIV Infections - diagnosis; HIV Infections - mortality; Humans; Infectious Disease; Late presentation; Male; Middle Aged; Mortality; Risk Factors; Spain - epidemiology; Survival Analysis; HIV; Causes of death; Mortality; Factors and trends; Cohort study; Late presentation
• ISSN: 0163-4453; 1532-2742
• DOI: 10.1016/j.jinf.2016.01.017

Summary Objectives To analyze the impact of late presentation (LP) on overall mortality and causes of death and describe LP trends and risk factors (2004–2013). Methods Cox models and logistic regression were used to analyze data from a nation-wide cohort in Spain. LP is defined as being diagnosed when CD4 < 350 cells/ml or AIDS. Results Of 7165 new HIV diagnoses, 46.9% (CI95% :45.7–48.0) were LP, 240 patients died. First-year mortality was the highest (aHRLP.vs.nLP  = 10.3[CI95% :5.5–19.3]); between 1 and 4 years post-diagnosis, aHRLP.vs.nLP  = 1.9(1.2–3.0); and >4 years, aHRLP.vs.nLP  = 1.5(0.7–3.1). First-year's main cause of death was HIV/AIDS (73%); and malignancies among those surviving >4 years (32%). HIV/AIDS-related deaths were more likely in LP (59.2% vs. 25.0%; p < 0.001). LP declined from 55.9% (2004–05) to 39.4% (2012–13), and reduced in 46.1% in men who have sex with men (MSM) and 37.6% in heterosexual men, but increased in 22.6% in heterosexual women. Factors associated with LP: sex (ORMEN.vs.WOMEN  = 1.4[1.2–1.7]); age (OR31–40.vs.<30  = 1.6[1.4–1.8], OR41–50.vs.<30  = 2.2[1.8–2.6], OR>50.vs.<30  = 3.6[2.9–4.4]); behavior (ORInjectedDrugUse.vs.MSM  = 2.8[2.0–3.8]; ORHeterosexual.vs.MSM  = 2.2[1.7–3.0]); education (ORPrimaryEducation.vs.University  = 1.5[1.1–2.0], ORLowerSecondary.vs.University  = 1.3[1.1–1.5]); and geographical origin (ORSub-Saharan.vs.Spain  = 1.6[1.3–2.0], ORLatin-American.vs.Spain  = 1.4[1.2–1.8]). Conclusions LP is associated with higher mortality, especially short-term- and HIV/AIDS-related mortality. Mid-term-, but not long-term mortality, remained also higher in LP than nLP. LP decreased in MSM and heterosexual men, not in heterosexual women. The groups most affected by LP are low educated, non-Spanish and heterosexual women.