Relationship of Transfusion and Iron-Related Complications to Cost of Care in Thalassemia.

Rationale: Determinants of the financial burdens of medical care for transfusion-dependent thalassemia syndromes (Thal major, TM) include lifelong transfusion (Tx), iron chelator therapy, screening for complications of iron overload/blood-borne infections, and treatment of complications.

Methods/Subjects: As a surrogate measure for medical care costs, we retrospectively categorized and tallied total hospital charges, inpatient and outpatient, over a two year period (Oct. 2001 to Sept 2003) at a single Thal treatment center, and in addition, per-patient deferoxamine (DFO) costs, using average wholesale price (AWP) as a proxy. Hospital financial records for all subjects followed routinely at the center were analyzed after removing identifiers. Thirty-one subjects with TM, and 8 with thal intermedia (TI, sporadically transfused) were assessed. Mean age was 20 +/− 12 y (SD), range 3–47 y. Charges were assessed as a function of age, Thal type (TI vs. TM), and by presence or absence of major Tx-related complications: diabetes (n=4), heart disease (n=7), and infection (HIV, n=1; or HCV, n=6). All subjects at the center had insurance for hospital and DFO charges: 22 private, 20 Medicaid, 2 medicare, 1 other, 7 in combination. One TM patient received erythrocytapheresis, 30 received simple transfusions to trough Hb ~ 9.5 g/dl.

Results: The 31 TM subjects generated $3.8 million in hospital charges over the two-year period. Mean charges were $61,701/yr per TM patient (Tx-related comprise 63% of charges; screening studies 7.6%, infection-related 1.8%, iron-related complications 16%, and other/not classifiable 11.6%). Complications were correlated with age (r = 0.71. p <.0001), and care was increasingly expensive with increasing numbers of complications (see table). As number of RBC units transfused and risk of complications both increase with age, multivariate modeling was used to assess the relationship of charges to iron-related complications (diabetes or heart disease, n=8), adjusting for age. Mean charges were $45,608/yr without, or $85,695/yr with either complication (p = 0.04). Age-adjusted charges were modestly higher for subjects with hepC or HIV (n=5) compared to those without ($87,500/yr vs. $61,500/yr, p=0.12). Annual DFO cost was assessed as (vials per day prescribed) X 365 X AWP ($14.96 per 500 mg vial), based on prescribed dose at the study mid-point. Mean # vials per day was 2.9, or $15,600 per year per TM patient. Mean total charges per TI patient were only $3,318/yr, or 5% of the mean for TM. A potential limitation of this analysis is that hospital charges and AWP are imperfect proxies for costs. However, precise cost data are difficult to ascertain or impute, and the present data are likely to reflect true financial costs of medical care within a power of two (i.e. charges/2 may be closer to true "cost"). Furthermore, non-medical costs, including lost days of school and work for patients and families; physiologic and psychosocial toll of disease and therapy are not included in these totals.

Conclusions: Costs of medical care in TM rise with age, based not only on higher transfusion costs and higher DFO use related to weight. Adjusting for age, iron-related complications are a major cost determinant and a likely proxy for poor chlelation over time. Improved chelator compliance could improve both complications and cost.