Updated Direct Costs of Medical Care for HIV-Infected Patients Within a Regional Population From 2006 to 2017

HB Krentz; Q Vu; MJ Gill


HIV Medicine. 2020;21(5):289-298. 

In This Article


Study Population

The number of SAC patients at year end increased from 982 in 2006 to 1813 in 2017 (Table 1). The proportion of male patients decreased from 80.2% to 73.6%; the proportion of Caucasians decreased from 69.6% to 52.1%. The median age increased from 43 to 48 years with the proportion > 50 years of age nearly doubling from 20.3% to 39.4%. A higher percentage of patients reported heterosexual sex as the most likely HIV transmission factor in 2017 than in 2006 (42.7% versus 27.9%, respectively), whereas injecting drug use (IDU) decreased from 16% to 8.5%. While the number of new patients diagnosed locally was < 100 per year, the proportion of 'late presenters' (i.e. < 350 cells/μL at diagnosis) remained persistently high, varying annually from 40.7% to 66.7% (Table 1). The proportion of patients with CD4 count > 500 cells/μL increased to 63.4%, while the percentage with suppressed viral load increased to 91.2% by 2017. The number of all-cause deaths was < 31 per year. The use of ART at SAC is shown in Table 1 and has been described in greater detail elsewhere[1] but generally reflects ARV choices and clinical practice in many developed world clinics, with decreases in the use of protease inhibitors (PIs) and nonnucleoside reverse transcriptase inhibitors (NNRTIs) and an increase in the use of integrase strand transfer inhibitors (INSTIs).

Reflecting the improved health of SAC patients, the mean number of clinic visits per year per patient decreased from 2.77 in 2006 to 2.52 in 2017 (Table 2) and the number of annual CD4 count and viral load tests performed declined from a mean of 3.22 and 3.38, respectively, in 2006 to 2.10 and 2.59, respectively, in 2017. In-patient hospitalizations, while variable, exhibited an overall declining trend after 2014 in both the number of patients hospitalized and the number of days admitted. HIV-related hospitalizations were significantly lower than non-HIV related admissions annually; non-HIV-related admissions accounted for the majority of all in-patient admissions.

Total Costs

The total cost of direct medical care for all HIV-infected patients within the region increased from $12.4 million in 2006 to $32.8 million in 2015, decreasing slightly to $30.1 million in 2017 (Figure 1). The cost of ARV drugs rose from $9.1 to $24 million and accounted for 74.4% of total costs, increasing from 73.3% to 78.8% by 2017. Annual out-patient visit and laboratory costs accounted for 9.4% of all costs but declined proportionally as part of all costs from 12% to 8.5% in 2017. In-patient hospitalization costs, especially non-HIV-related admissions, showed annual variability, accounting for 13.3% of all costs overall but increasing from 9.9% in 2006 to 15.7% in 2014 to 8.6% in 2017. HIV-related admissions ranged from 1.9% to 4.2% of all costs, whereas non-HIV-related admissions ranged from 7.8% to 11.5% of all costs.

Figure 1.

Total cost of care for all patients followed at the Southern Alberta Clinic (SAC) subdivided by costing category in 2017 Cdn$.

PPPM Costs

The mean (SE) total PPPM cost of care increased from $1316 ($256) in 2006 to $1712 ($311) in 2014, declining to $1446 ($277) in 2017 (Table 3). Higher PPPM costs were associated with CD4 counts < 200 cells/μL. Although patients with CD4 counts < 75 cells/μL continued to exhibit the highest costs, the mean PPPM cost for patients with CD4 counts > 75 cells/μL became more homogenous and overlapping over time, making CD4 count less useful as a costing category than in previous studies. Higher costs at lower CD4 counts were mostly related to hospital admissions. In-patient care costs PPPM ranged from $417 (±$233) in 2015 to $2565 (±$1117) in 2007 for patients with CD4 counts < 75 cells/μL. The variability in annual PPPM costs was associated with a small number of patients with very high in-patient costs as a consequence of serious HIV- and non-HIV-related conditions. The mean PPPM cost for patients with higher CD4 counts has increased as more patients are placed and remain on ARV drugs, with their ARVs comprising a greater proportion of total costs of care. Out-patient care costs PPPM declined for all CD4 count categories. The mean PPPM cost for HIV-related admissions for patients with CD4 counts > 200 cells/μL was relatively small (although the cost of a single hospital admission may be large), with non-HIV-related admission accounting for most in-patient costs.

HIV-related Costs Only

The mean PPPM costs for HIV-related costs (i.e. ARV drugs, HIV-related out-patient visits and HIV-related in-patient visits only) remained relatively stable for patients with a CD4 count > 200 cells/μL (Figure 2), increasing by < 10% per year. PPPM costs varied more significantly for patients with CD4 counts < 200 cells/μL, with most of the variation attributable to in-patient admissions, which can vary greatly from year to year.

Figure 2.

Mean cost per patient per month (PPPM) for HIV-related costs [i.e. antiretroviral (ARV) drugs, HIV clinic and laboratory tests, and HIV-related hospitalizations] in 2017 Cdn$. CD4 categories are cells/μL.

Cost Comparisons Between 2005 and 2017

Using mean PPPM cost as the metric, we compared medical costs for patients in our population in 2005 and 2017 adjusting to 2017 Cdn $ equivalency (Figure 3). Although the total costs of care for the entire population increased significantly, the mean PPPM costs increased by only 5%, from $1379 in 2005 to $1446 in 2017. Costs, however, shifted as a consequence of practice change and treatment effects. The mean PPPM cost of ARV drugs increased by 32%, from $863 to $1139, but the PPPM cost for out-patient and in-patient services decreased by 80% and 71%, respectively. Most of the decrease for in-patient costs was attributable to a substantial decrease in HIV-related hospitalizations.

Figure 3.

Comparison of mean costs per patient per month (PPPM) between 2005 and 2017 in 2017 Cdn $. [2005 data from Krentz et al. 28].