A MultiFactorial Risk Score to weigh toxicities and co‐morbidities relative to costs of antiretrovirals in a cohort of HIV‐infected patients

Purpose of the study Considering costs of antiretrovirals (ARVs) for HIV patients is increasingly needed. A simple and comprehensive tool weighing comorbidities and ARV‐related toxicities could be useful to judge the appropriateness of use of more expensive drugs. We conceived a MultiFactorial Risk Score (MFRS) to evaluate the appropriateness of ARVs prescription relative to their costs. Methods HIV patients were consecutively enrolled in 2010‐2011. We considered socio‐demographic characteristics, HIV history, cardiovascular risk factors, low energy fractures, bone density. Psychological factors were assessed by BDI, DS14 and TAS‐20. The MFRS was calculated as the sum of the following: age (<30y 1 point; 1 point increase every 5y, 10 for≥70); AIDS diagnosis (5); CD4 nadir (5 if <100; 1 point less every 100 CD4 increase); ART line (0 first, up to 5 for≥6 lines); lipodistrophy (5); HCV coinfection (7); education (1 degree, 2 secondary, 3 primary); alcohol (3) and drug abuse (5); working activity (3 if unemployed); hypertension (3); cholesterol≥200 mg/dl (3); diabetes (3); Framingham score (7 if>7%); creatinine (0 if <1 mg/dl, 1 if<1.2; 2 if<1.5>1.2, 5 if<2> 1.5, 7 if≥2); bone fractures (7); bone status at DEXA (0 normal, 3 osteopenic, 5 osteoporotic); cancer (5); depression (3 if BDI>17); other psychiatric illness (5). Annual costs of individual ART regimens were calculated. MFRS was correlated in univariate and multivariate models with all variables. All statistical analyses were carried out using Stata 10.1. Summary of results We enrolled 241 HIV patients, 74.3% males, aged 44.5±9.9y; 19 patients (7.8%) were untreated, 74.8% of treated had undetectable HIV RNA. Mean Nadir CD4 counts were 218±168, 38.5% of patients had an AIDS diagnosis. Mean individual ARV annual cost was 10,976±5,360. Mean MFRS was 28.5±13.9 (4–64). MFRS was significantly higher (p<0.001) in patients with older age, longer duration of HIV infection, lower CD4 nadirs, AIDS diagnosis, lipodistrophy, HCV, smoking, lower education, alcohol/drug abuse, hypertension, carotid plaques, higher Framingham score, diabetes, bone fractures or disorders, depression, alexithymia, and higher ARV costs. In multivariate models, ARV costs were significantly higher in patients with older age, previous AIDS diagnosis, lower CD4 nadir and higher MFRS. Conclusions MFRS may be a simple and reliable tool to match patients’ complexity and ARV costs, deserving further validation on larger samples.