Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

The Importance of Clinical Variables in Comparative Analyses Using Propensity-Score Matching

The Case of ESA Costs for the Treatment of Chemotherapy-Induced Anaemia

  • 75 Accesses

  • 14 Citations



The erythropoiesis-stimulating agents (ESAs) epoetin alfa (EA) and darbepoetin alfa (DA) have comparable efficacy in treating chemotherapy-induced anaemia (CIA). Therapy choice depends on many factors, including cost. Previous estimates of ESA cost differences have been derived from claims data. These data lack clinical variables, such as baseline haemoglobin (Hb) level, which are likely to influence choice of ESA, dosing and costs. We estimated cost differences between DA and EA in patients with cancer receiving chemotherapy, using a propensity-score matched analysis of baseline patient characteristics with and without Hb values to assess the effect of this clinical variable on ESA cost estimates.


Data were extracted from electronic medical records in two US databases between January 2004 and December 2006. The study sample included 6743 patients receiving chemotherapy, with one or more visits during the study period, who received an ESA during a chemotherapy episode. Episodes of chemotherapy care were constructed using a 90-day gap in administration to identify the start and end. Patients receiving both DA and EA during their initial chemotherapy episode or with missing data were excluded, representing 42% of patients with CIA receiving an ESA. Drug costs were calculated from the cumulative dose multiplied by 106% of the average sales price (ASP) for DA or EA. Two propensity-score matches were conducted: first using variables available in administrative billing claims systems, then adding the baseline Hb test result. Regression-adjusted cost differences were estimated with and without baseline Hb, using generalized linear models.


Using baseline Hb levels resulted in a better match of the baseline characteristics for the EA and DA treatment groups than the original sample or the matched sample without Hb variables. Mean ESA costs (year 2007 values) for the original sample were $US4171 for EA and $US3811 for DA (mean difference $US360; p<0.001, standard error [SE] $US99). With propensity-score matching without Hb variables, mean estimated costs were $US3836 for EA and $US3599 for DA (mean difference $US237; p = 0.053, SE $US123). With propensity-score match including Hb variables, mean costs were $US3965 for EA and $US3536 for DA (mean difference $US429; p = 0.001, SE $US125). Cost differences in sensitivity analyses ranged between $US102 (p = 0.201) and $US261 (p = 0.003).


Addition of baseline Hb level as a variable in propensity score and ESA cost models affects ESA treatment cost estimates in patients with cancer receiving chemotherapy. Cost comparisons based on observational data should use analytical methods that account for differences in clinical variables between treatment groups.

This is a preview of subscription content, log in to check access.


  1. 1.

    D’Agostino Jr RB. Propensity score methods for bias reduction in the comparison of a treatment to a nonrandomized control group. Stat Med 1998; 17: 2265–81

  2. 2.

    Groeneveld PW, Matta MA, Greenhut AP, et al. The costs of drug-eluting coronary stents among Medicare beneficiaries. Am Heart J 2008; 155 (6): 1097–105

  3. 3.

    Ye X, Sikirica V, Schein JR, et al. Treatment failure rates and health care utilization and costs among patients with community-acquired pneumonia treated with levofloxacin or macrolides in an outpatient setting: a retrospective claims database analysis. Clin Ther 2008; 30 (2): 358–71

  4. 4.

    Huskamp HA, Deverka PA, Landrum MB, et al. The effect of three-tier formulary adoption on medication continuation and spending among elderly retirees. Health Serv Res 2007; 42 (5): 1926–42

  5. 5.

    Wu E, Chen L, Birnbaum H, et al. Cost of care for patients with rheumatoid arthritis receiving TNF-antagonist therapy using claims data. Curr Med Res Opin 2007; 23 (8): 1749–59

  6. 6.

    Baser O. Too much ado about propensity score models? Comparing methods of propensity score matching. Value Health 2006; 9 (6): 377–85

  7. 7.

    National Comprehensive Cancer Network. Anemia Panel. NCCN clinical practice guidelines in oncology: cancer- and treatment-related anemia, version V0.3.2007 [online]. Available from URL: [Accessed 2007 Dec 3]

  8. 8.

    Boccia R, Malik IA, Raja V, et al. Darbepoetin alfa administered every three weeks is effective for the treatment of chemotherapy-induced anemia. Oncologist 2006; 11:409–17

  9. 9.

    Waltzman R, Croot C, Justice GR, et al. Randomized comparison of epoetin alfa (40 000 U weekly) and darbepoetin alfa (200 microg every 2 weeks) in anemic patients with cancer receiving chemotherapy. Oncologist 2005; 10: 642–50

  10. 10.

    Glaspy J, Vadhan-Raj S, Patel R, et al. Randomized comparison of every-2-week darbepoetin alfa and weekly epoetin alfa for the treatment of chemotherapy-induced anemia: the 20030125 Study Group Trial. J Clin Oncol 2006; 24: 2290–7

  11. 11.

    Ross SD, Allen IE, Henry DH, et al. Clinical benefits and risks associated with epoetin and darbepoetin in patients with chemotherapy-induced anemia: a systematic review of the literature. Clin Ther 2006; 28: 801–31

  12. 12.

    Reed SD, Radeva JI, Daniel DB, et al. Economic evaluation of weekly epoetin alfa versus biweekly darbepoetin alfa for chemotherapy-induced anaemia: evidence from a 16-week randomised trial. Pharmacoeconomics 2006; 24: 479–94

  13. 13.

    Daniel GW, Hurley D, Whyte J, et al. Analysis of dose, cost and use patterns of erythropoietin stimulating agents in cancer patients [abstract]. Value Health 2007; 10: A134

  14. 14.

    Berger A, Edelsberg J, Kallich JD, et al. Use of darbepoetin alfa and epoetin alfa in clinical practice in patients with cancer-induced anemia. Clin Ther 2008; 30: 206–18

  15. 15.

    Luo W, Nordstrom B, Ranganathan G, et al. Adherence to guidelines for use of erythropoiesis stimulating proteins in patients with chemotherapy-induced anemia: trends from electronic medical records. Value Health 2007; 10: A136–7

  16. 16.

    Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45: 613–9

  17. 17.

    Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for casual effects. Biometrika 1983; 70: 41–55

  18. 18.

    Rosenbaum PR, Rubin DB. Reducing bias in observational studies using subclassification on the propensity score. J Am Stat Assoc 1984; 79: 516–24

  19. 19.

    Manning WG, Mullahy J. Estimating log models: to transform or not to transform? J Health Econ 2001; 20: 461–94

  20. 20.

    Leuven E, Sianesi B. PSMATCH2: STATA module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing [online]. Available from URL: [Accessed 2007 Dec 3]

  21. 21.

    Centers for Medicare and Medicaid Services. Decision memo for erythropoiesis stimulating agents (ESAs) for non-renal disease indications (CAG-00383N) [online]. Available from URL: [Accessed 2007 Jul 30]

Download references

Author information

Correspondence to Dr Jerrold Hill.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Polsky, D., Eremina, D., Hess, G. et al. The Importance of Clinical Variables in Comparative Analyses Using Propensity-Score Matching. Pharmacoeconomics 27, 755–765 (2009).

Download citation


  • Propensity Score
  • Cost Difference
  • Darbepoetin Alfa
  • Matched Sample
  • Average Sales Price