In the era of HAART, the prevalence of HIV-associated neuropsychiatric disorders (NPD) continues to rise , in part because of longer survival times. Currently, the two principal HIV-related NPD found in clinics in the industrialized world are cognitive impairments (CI) and peripheral neuropathies (PN), which are chiefly defined by subcortical neuropsychiatric dysfunction and distal limb neuropathic pain, respectively . Indeed, both groups of disorders are more frequently encountered after the development of AIDS . CI frequently necessitates extensive investigations to exclude other disorders and eventually requires institutional care. PN, caused by HIV infection, is associated with intercurrent diseases such as diabetes mellitus as well as the neurotoxic effects of some antiretroviral therapy (ART) and also requires ongoing medical management .
Since the implementation of HAART in the mid-1990s, increasing healthcare costs reflect the changes in treatment strategies, including the use of more antiretroviral medications. Previous costing studies have examined the primary healthcare costs for patients with HIV/AIDS at different stages of infection [5–7]. However, few studies have examined recent costing trends taking into account the impact of comorbid (neuropsychiatric) diseases. The present study examines the hypothesis that NPD would increase the costs of care among patients with HIV/AIDS, given the high costs of managing NPD.
The Southern Alberta Clinic is a centralized, multidisciplinary clinic that provides all outpatient healthcare services to patients with HIV/AIDS in southern Alberta. This includes clinical care at regular intervals (3–4 months), laboratory investigations (CD4 and CD8 lymphocyte counts, viral loads, etc.), ART and related medications, as well as specialty/subspecialty services, all without financial cost to the patient [8,9]. A database was developed at the clinic to collect direct costing information for each patient and to provide prospective clinical, epidemiological, socioeconomic and outcome data . The database has been utilized in studies approved by the Conjoint Medical Ethics Committee (University of Calgary).
All HIV-seropositive patients who received a diagnosis of NPD between January 1997 and March 2003 were included. Diagnoses were made by the clinic neurologist (C.P.) using established clinical criteria . Patients were subsequently stratified into three groups: (a) HIV-related CI, including HIV-associated dementia or minor cognitive motor disorder; (b) HIV-related PN, including distal sensory polyneuropathy or antiretroviral toxic peripheral neuropathy; and (c) other neuropsychiatric disorders (OND), including stroke/transient ischemic attacks, seizure, headache, or mononeuropathy that developed after the diagnosis of HIV infection.
If more than one concurrent NPD was identified, the initial diagnosis was used to classify the patient. Those patients who demonstrated severe cognitive impairment (Memorial Sloan Kettering dementia severity score ≥ 2)  were excluded from the present analysis. HIV-positive patients without NPD (non-NPD) were selected from the same relational database by matching the NPD group based on prognostic indicators of HIV disease: age, duration of HIV infection, CD4 cell count at diagnosis, nadir CD4 cell count, peak viral load, ethnicity, education, HIV infection risk factor, concurrent infections, substance abuse, mortality and the absence of an NPD.
Direct costs were collected for a number of factors: ART and non-antiretroviral drugs, laboratory, clinic, primary care and specialty physician visits, HIV and non-HIV-related hospitalizations, and homecare costs, but not neuroimaging costs. Interval costs for each category were averaged and expressed as mean cost per patient per month (CPM) in 2003 Canadian dollars (±SEM). Interval data were obtained for the three neuropsychiatric groups up to 24 months prior to the neuropsychiatric diagnosis and grouped as 18–24, 12–18, and up to 12 months prior to diagnosis, while the CPM over the entire duration of the study was used for the non-NPD group. Following diagnosis of a NPD, data were obtained at 6-month intervals for the available months until an assigned end date of 31 March 2003. Patients who moved or died prior to this date were included up the time they were lost to follow-up. Only patients with costs for ≥ 2 months were included in the analysis. Statistical analyses using parametric and non-parametric tests were performed using Graph-Pad Instat version 3.0 (Graph Pad software, San Diego, California, USA) and P values < 0.05 were considered significant.
The study included 188 HIV-1-seropositive patients with a neuropsychiatric diagnosis and 153 NPD-free patients. The non-NPD group was selected based on matching disease indicators, which did not differ significantly among clinical groups: mean values for non-NPD and NPD groups, respectively, were 44.7 and 47.1 years in age, 4097 and 4214 days for duration of HIV infection, 343 and 379 cells/μl for CD4 cell count at diagnosis, 139 and 131 cells/μl for nadir CD4 cell count, and 5.3 and 5.8 log10 copies/ml for peak viral load. In addition to these variables, other sociodemographic and clinical parameters including sex, years of education, risk factor for HIV infection (men who have sex with men, intravenous drug use, other), ethnicity (percentage Caucasian), hepatitis B and C virus infection, syphilis, substance abuse, and mortality were not significantly different between the NPD and non-NPD groups (data not shown). The NPD group was further stratified into subgroups: 38 with CI, 73 with PN, and 77 with OND. Again, there were no significant differences in the above variables among individual subgroups or relative to the non-NPD group.
Comparison of total mean CPM of all cost variables after diagnosis of NPD (Table 1) revealed significant differences between the non-NPD control group ($916.03 ± 58.21) and the CI ($1612.52 ± 195.93), PN ($1489.91 ± 99.02), and OND groups ($1361.85 ± 90.11) (P < 0.001). For all patients with NPD, increased mean CPM was generated by additional clinic visits (P < 0.001), physician visits (P < 0.001), and non-antiretroviral drugs (P < 0.01). However, there were also notable group-specific differences. The CI group had significantly increased requirements for homecare (P < 0.05), associated with a greater degree of impairment for patients with CI, incurring over 10 times the cost of the control group (mean CPM $103.09 ± 39.98 and $10.42 ± 4.92, respectively). The PN and OND groups each exhibited post-diagnosis increases in ART costs [mean CPM $908.02 ± 63.19 (P < 0.01) and $850.43 ± 54.21 (P < 0.05), respectively].
Comparison of total mean CPM for groups at 12 months before diagnosis of a NPD (Fig. 1a) also revealed a significant difference between non-NPD controls ($916.03 ± 58.21) and the entire NPD group ($1371.06 ± 97.55) (P < 0.001). Analysis of the individual costs revealed that for the CI group there was increased expenditure for clinic visits (CPM $135.67 ± 13.15; P < 0.01) and physician visits (CPM $30.71 ± 6.04; P < 0.001) compared with the non-NPD group. The PN and OND groups were similar, incurring increased costs not only for clinic visits [CPM $156.86 ± 15.23 and $150.19 ± 12.11, respectively (P < 0.001)] and physician visits [CPM $18.07 ± 1.74 and $17.21 ± 2.27, respectively (P < 0.001)] but also for laboratory costs [CPM $62.30 ± 4.72 and $63.33 ± 5.52, respectively (P < 0.05] and specialty physician visits [CPM $10.69 ± 1.12 and $13.28 ± 1.80, respectively (P < 0.001)]. In addition, patients with PN also had higher costs for non-antiretroviral drugs (CPM $77.97 ± 13.80; P < 0.001). Notably, there were no significant differences in mean CPM for ART (P > 0.05) comparing the NPD groups with the non-NPD group.
Data were also collected to investigate mean CPM costs at the 12–18 and 18–24 month intervals prior to and after diagnosis (Fig. 1b). Total mean CPM for patients with NPD from the 12–18 month interval pre-diagnosis ($1060.67 ± 65.69) and the 18–24 month interval pre-diagnosis ($1019.32 ± 54.09) did not differ significantly from the non-NPD group (P > 0.05). The total mean overall costs over time for the CI and PN groups showed a sharp rise in costs after diagnoses, which subsequently stabilized. In contrast, the OND group showed a minimal increase in overall costs after diagnosis (Fig. 1b). Analysis of the total mean CPM also revealed that patients with two or more neuropsychiatric diagnoses cost significantly more than those with only one diagnosis [$1659.59 ± 101.89 and $1368.12 ± 85.47, respectively (P < 0.05].
The present study represents the first cost-of-care analysis for patients with HIV/AIDS and NPD in the era of HAART. In our study, patients with HIV/AIDS and a diagnosis of NPD were more costly both immediately before and after diagnosis than HIV-positive patients without a NPD when matched for age, sex, and systemic HIV disease, regardless of the specific neuropsychiatric diagnosis. Indeed, costs appeared to ‘anticipate’ the diagnosis of a NPD by over 1 year in a prospective analysis. HIV-associated CI showed the greatest increases in costs, particularly in terms of the expenses of physician and clinic visits. Following diagnosis, homecare costs also rose significantly (over 10-fold higher than control), particularly for those patients with HIV-associated dementia (over 14-fold), underscoring the increased healthcare needs in this group.
All patient groups with a diagnosis of NPD exhibited higher costs for clinic visits, physician (clinic and specialty) visits, and laboratory visits at least 12 months prior to diagnosis. Importantly, ART costs did not rise prior to diagnosis confirmation, implying that there was no change in HIV status during this period. Hence, these higher costs may reflect the result of the slow progression of symptoms with accompanying costs of investigation before a formal diagnosis is made, or under-recognized existing disease . Following diagnosis, ongoing high costs are likely indicative of services needed to manage a greater NPD-related disease burden. Furthermore, non-ART medication costs increased for NPD groups after diagnosis, which suggests that concomitant neurological disease adds more vulnerability to HIV/AIDS-associated illness or medication efficacy.
Not unexpectedly, the presence of more than one neuropsychiatric diagnosis further increased costs, emphasizing the high cost burden of NPD during HIV infection [2,12]. Moreover, analysis of data for up to 18 months after diagnosis indicated that increased costs for concurrent NPD were sustained longitudinally but did not continue to rise. The study, therefore, emphasizes the substantially increased costs of care for patients with NPD, particularly individuals with CI.
Our findings are particularly relevant in the context of extended survival of patients in the HAART era, as NPD is one of the major sources of morbidity and mortality in patients with HIV/AIDS . In part, the difference between the study described here and previous studies in terms of ART costs [7,14] may reflect the specific disease stages of patients, with our groups exhibiting more advanced disease, as indicated by longer durations of infection and lower CD4 cell count nadirs. Nonetheless, because our study was conducted in a regionally based community clinic, the findings are representative of a broad HIV/AIDS-positive population, which may also influence the relative costs of patients' care management, as different sources of costs were used for analysis, including hospital costs, outpatient costs, and the costs of long-term institutionalization. We predict that health costs arising from NPD are likely to rise as the use of HAART generates a patient population with HIV/AIDS that has extended life expectancy, necessitating health care planning for these costs. In future studies, it might be useful to examine a wider range of NPDs together with the impact of other intercurrent illnesses, including other infections, substance abuse, and other health determinants that influence healthcare costs.
We thank Michael Henry for his assistance in data collection, and Sherry Sweeney for assistance with manuscript preparation. C.P. holds a Canada Research Chair (T1) in Neurological Infection and Immunity.
1. McArthur JC, Brew BJ, Nath A. Neurological complications of HIV infection. Lancet Neurol 2005; 4:543–555.
2. Sperber K, Shao L. Neurologic consequences of HIV infection in the era of HAART. AIDS Patient Care STD 2003; 17:509–518.
3. Power C, Gill MJ, Johnson RT. The neuropathogenesis of HIV infection: host-virus interaction and the impact of therapy. Can J Neurol Sci 2002; 29:19–32.
4. Youle M. HIV-associated antiretroviral toxic neuropathy (ATN): a review of recent advances in pathophysiology and treatment. Antivir Ther 2005; 10(Suppl 2):M125–M129.
5. Anis AH, Guh D, Hogg RS, Wang XH, Yip B, Craib KJ, et al
. The cost effectiveness of antiretroviral regimens for the treatment of HIV/AIDS. Pharmacoeconomics 2000; 18:393–404.
6. Gebo KA, Chaisson RE, Folkemer JG, Bartlett JG, Moore RD. Costs of HIV medical care in the era of highly active antiretroviral therapy. AIDS 1999; 13:963–969.
7. Bozzette SA, Joyce G, McCaffrey DF, Leibowitz AA, Morton SC, Berry SH, et al
. Expenditures for the care of HIV-infected patients in the era of highly active antiretroviral therapy. N Engl J Med 2001; 344:817–823.
8. Krentz HB, Auld MC, Gill MJ. The changing direct costs of medical care for patients with HIV/AIDS, 1995–2001. Can Med Assoc J 2003; 169:106–110.
9. Krentz HB, Auld MC, Gill MJ. The high cost of medical care for patients who present late (CD4 < 200 cells/microl) with HIV infection. HIV Med 2004; 5:93–98.
10. Janssen R. Nomenclature and research case definitions for neurologic manifestations of human immunodeficiency virus-type 1 (HIV-1) infection. Report of a Working G roup of the American Academy of Neurology AIDS Task Force [see comments]. Neurology 1991; 41:778–785.
11. Price RW, Brew BJ. The AIDS dementia complex. J Infect Dis 1988; 158:1079–1083.
12. Dal Pan GJ, Skolasky RL, Moore RD. The impact of neurologic disease on hospitalisations related to human immunodeficiency virus-infection in Maryland 1991–1992. Arch Neurol 1997; 54:846–852.
13. Sacktor N. The epidemiology of human immunodeficiency virus-associated neurological disease in the era of highly active antiretroviral therapy. J Neurovirol 2002; 8(Suppl 2):115–121.
14. Gable CB, Tierce JC, Simison D, Ward D, Motte K. Costs of HIV+/AIDS at CD4+ counts disease stages based on treatment protocols. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12:413–420.