Korean Journal of Radiology

The Korean Society of Radiology (대한영상의학회)
권호 표지
DOI QR코드

DOI QR Code

Propensity Score Matching: A Conceptual Review for Radiology Researchers

  • Baek, Seunghee (Department of Clinical Epidemiology and Biostatistics, Asan Medical Center) ;
  • Park, Seong Ho (Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine) ;
  • Won, Eugene (Department of Radiology, NYU Langone Medical Center) ;
  • Park, Yu Rang (Office of Clinical Research Information, Asan Medical Center) ;
  • Kim, Hwa Jung (Department of Clinical Epidemiology and Biostatistics, Asan Medical Center)
  • Received : 2014.07.08
  • Accepted : 2014.11.28
  • Published : 2015.04.01
⟨ Previous Next ⟩

Abstract

The propensity score is defined as the probability of each individual study subject being assigned to a group of interest for comparison purposes. Propensity score adjustment is a method of ensuring an even distribution of confounders between groups, thereby increasing between group comparability. Propensity score analysis is therefore an increasingly applied statistical method in observational studies. The purpose of this article was to provide a step-by-step nonmathematical conceptual guide to propensity score analysis with particular emphasis on propensity score matching. A software program code used for propensity score matching was also presented.

Keywords

References

  1. Psaty BM, Siscovick DS. Minimizing bias due to confounding by indication in comparative effectiveness research: the importance of restriction. JAMA 2010;304:897-898 https://doi.org/10.1001/jama.2010.1205
  2. Primrose JN, Perera R, Gray A, Rose P, Fuller A, Corkhill A, et al. Effect of 3 to 5 years of scheduled CEA and CT follow-up to detect recurrence of colorectal cancer: the FACS randomized clinical trial. JAMA 2014;311:263-270 https://doi.org/10.1001/jama.2013.285718
  3. Kim K, Kim YH, Kim SY, Kim S, Lee YJ, Kim KP, et al. Lowdose abdominal CT for evaluating suspected appendicitis. N Engl J Med 2012;366:1596-1605 https://doi.org/10.1056/NEJMoa1110734
  4. Trinchet JC, Chaffaut C, Bourcier V, Degos F, Henrion J, Fontaine H, et al. Ultrasonographic surveillance of hepatocellular carcinoma in cirrhosis: a randomized trial comparing 3- and 6-month periodicities. Hepatology 2011;54:1987-1997 https://doi.org/10.1002/hep.24545
  5. Fischer B, Lassen U, Mortensen J, Larsen S, Loft A, Bertelsen A, et al. Preoperative staging of lung cancer with combined PETCT. N Engl J Med 2009;361:32-39 https://doi.org/10.1056/NEJMoa0900043
  6. Righini M, Le Gal G, Aujesky D, Roy PM, Sanchez O, Verschuren F, et al. Diagnosis of pulmonary embolism by multidetector CT alone or combined with venous ultrasonography of the leg: a randomised non-inferiority trial. Lancet 2008;371:1343-1352 https://doi.org/10.1016/S0140-6736(08)60594-2
  7. Rosenberger WF, Lachin JM. Randomization and the clinical trial. In: Rosenberger WF, Lachin JM, eds. Randomization in clinical trials: theory and practice, 1st ed. New York: Wiley- Interscience, 2002:1-14
  8. Cha DI, Lee MW, Rhim H, Choi D, Kim YS, Lim HK. Therapeutic efficacy and safety of percutaneous ethanol injection with or without combined radiofrequency ablation for hepatocellular carcinomas in high risk locations. Korean J Radiol 2013;14:240-247 https://doi.org/10.3348/kjr.2013.14.2.240
  9. Chung SY, Park SH, Lee SS, Lee JH, Kim AY, Park SK, et al. Comparison between CT colonography and double-contrast barium enema for colonic evaluation in patients with renal insufficiency. Korean J Radiol 2012;13:290-299 https://doi.org/10.3348/kjr.2012.13.3.290
  10. Kim DH, Pickhardt PJ, Taylor AJ, Leung WK, Winter TC, Hinshaw JL, et al. CT colonography versus colonoscopy for the detection of advanced neoplasia. N Engl J Med 2007;357:1403-1412 https://doi.org/10.1056/NEJMoa070543
  11. Kim JW, Shin SS, Kim JK, Choi SK, Heo SH, Lim HS, et al. Radiofrequency ablation combined with transcatheter arterial chemoembolization for the treatment of single hepatocellular carcinoma of 2 to 5 cm in diameter: comparison with surgical resection. Korean J Radiol 2013;14:626-635 https://doi.org/10.3348/kjr.2013.14.4.626
  12. Lee SH, Chung CH, Jung SH, Lee JW, Shin JH, Ko KY, et al. Midterm outcomes of open surgical repair compared with thoracic endovascular repair for isolated descending thoracic aortic disease. Korean J Radiol 2012;13:476-482 https://doi.org/10.3348/kjr.2012.13.4.476
  13. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41-55 https://doi.org/10.1093/biomet/70.1.41
  14. Choi GH, Shim JH, Kim MJ, Ryu MH, Ryoo BY, Kang YK, et al. Sorafenib alone versus sorafenib combined with transarterial chemoembolization for advanced-stage hepatocellular carcinoma: results of propensity score analyses. Radiology 2013;269:603-611 https://doi.org/10.1148/radiol.13130150
  15. McDonald JS, McDonald RJ, Fan J, Kallmes DF, Lanzino G, Cloft HJ. Comparative effectiveness of ruptured cerebral aneurysm therapies: propensity score analysis of clipping versus coiling. AJNR Am J Neuroradiol 2014;35:164-169 https://doi.org/10.3174/ajnr.A3642
  16. McDonald JS, Kallmes DF, Lanzino G, Cloft HJ. Percutaneous closure devices do not reduce the risk of major access site complications in patients undergoing elective carotid stent placement. J Vasc Interv Radiol 2013;24:1057-1062 https://doi.org/10.1016/j.jvir.2013.03.030
  17. McDonald RJ, McDonald JS, Bida JP, Carter RE, Fleming CJ, Misra S, et al. Intravenous contrast material-induced nephropathy: causal or coincident phenomenon? Radiology 2013;267:106-118 https://doi.org/10.1148/radiol.12121823
  18. Davenport MS, Khalatbari S, Cohan RH, Dillman JR, Myles JD, Ellis JH. Contrast material-induced nephrotoxicity and intravenous low-osmolality iodinated contrast material: risk stratification by using estimated glomerular filtration rate. Radiology 2013;268:719-728 https://doi.org/10.1148/radiol.13122276
  19. Davenport MS, Khalatbari S, Dillman JR, Cohan RH, Caoili EM, Ellis JH. Contrast material-induced nephrotoxicity and intravenous low-osmolality iodinated contrast material. Radiology 2013;267:94-105 https://doi.org/10.1148/radiol.12121394
  20. Takuma Y, Takabatake H, Morimoto Y, Toshikuni N, Kayahara T, Makino Y, et al. Comparison of combined transcatheter arterial chemoembolization and radiofrequency ablation with surgical resection by using propensity score matching in patients with hepatocellular carcinoma within Milan criteria. Radiology 2013;269:927-937 https://doi.org/10.1148/radiol.13130387
  21. de Haan MC, Boellaard TN, Bossuyt PM, Stoker J. Colon distension, perceived burden and side-effects of CTcolonography for screening using hyoscine butylbromide or glucagon hydrochloride as bowel relaxant. Eur J Radiol 2012;81:e910-e916 https://doi.org/10.1016/j.ejrad.2012.05.020
  22. McDonald RJ, McDonald JS, Kallmes DF, Carter RE. Behind the numbers: propensity score analysis-a primer for the diagnostic radiologist. Radiology 2013;269:640-645 https://doi.org/10.1148/radiol.13131465
  23. Lee J, Cho JY, Lee HJ, Jeong YY, Kim CK, Park BK, et al. Contrast-induced nephropathy in patients undergoing intravenous contrast-enhanced computed tomography in Korea: a multi-institutional study in 101487 patients. Korean J Radiol 2014;15:456-463 https://doi.org/10.3348/kjr.2014.15.4.456
  24. Altman DG. The scandal of poor medical research. BMJ 1994;308:283-284 https://doi.org/10.1136/bmj.308.6924.283
  25. Salas M, Hofman A, Stricker BH. Confounding by indication: an example of variation in the use of epidemiologic terminology. Am J Epidemiol 1999;149:981-983 https://doi.org/10.1093/oxfordjournals.aje.a009758
  26. Sica GT. Bias in research studies. Radiology 2006;238:780-789 https://doi.org/10.1148/radiol.2383041109
  27. Gunderman RB. Biases in radiologic reasoning. AJR Am J Roentgenol 2009;192:561-564 https://doi.org/10.2214/AJR.08.1220
  28. Ladapo JA, Blecker S, Elashoff MR, Federspiel JJ, Vieira DL, Sharma G, et al. Clinical implications of referral bias in the diagnostic performance of exercise testing for coronary artery disease. J Am Heart Assoc 2013;2:e000505 https://doi.org/10.1161/JAHA.113.000505
  29. Rubin DB, Thomas N. Matching using estimated propensity scores: relating theory to practice. Biometrics 1996;52:249-264 https://doi.org/10.2307/2533160
  30. Brookhart MA, Schneeweiss S, Rothman KJ, Glynn RJ, Avorn J, Sturmer T. Variable selection for propensity score models. Am J Epidemiol 2006;163:1149-1156 https://doi.org/10.1093/aje/kwj149
  31. Yang HJ, Lee JH, Lee DH, Yu SJ, Kim YJ, Yoon JH, et al. Small single-nodule hepatocellular carcinoma: comparison of transarterial chemoembolization, radiofrequency ablation, and hepatic resection by using inverse probability weighting. Radiology 2014;271:909-918 https://doi.org/10.1148/radiol.13131760
  32. Halpern EF. Behind the numbers: inverse probability weighting. Radiology 2014;271:625-628 https://doi.org/10.1148/radiol.14140035
  33. Kurth T, Walker AM, Glynn RJ, Chan KA, Gaziano JM, Berger K, et al. Results of multivariable logistic regression, propensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect. Am J Epidemiol 2006;163:262-270 https://doi.org/10.1093/aje/kwj047
  34. McAfee AT, Ming EE, Seeger JD, Quinn SG, Ng EW, Danielson JD, et al. The comparative safety of rosuvastatin: a retrospective matched cohort study in over 48,000 initiators of statin therapy. Pharmacoepidemiol Drug Saf 2006;15:444- 453 https://doi.org/10.1002/pds.1281
  35. Austin PC. A critical appraisal of propensity-score matching in the medical literature between 1996 and 2003. Stat Med 2008;27:2037-2049 https://doi.org/10.1002/sim.3150
  36. Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011;10:150-161 https://doi.org/10.1002/pst.433
  37. Austin PC. Propensity-score matching in the cardiovascular surgery literature from 2004 to 2006: a systematic review and suggestions for improvement. J Thorac Cardiovasc Surg 2007;134:1128-1135 https://doi.org/10.1016/j.jtcvs.2007.07.021
  38. Gu XS, Rosenbaum PR. Comparison of multivariate matching methods: structures, distances, and algorithms. J Comput Graph Stat 1993;2:405-420
  39. Ho DE, Imai K, King G, Stuart EA. Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political Analysis 2007;15:199-236 https://doi.org/10.1093/pan/mpl013
  40. Hill J. Discussion of research using propensity-score matching: comments on 'A critical appraisal of propensityscore matching in the medical literature between 1996 and 2003' by Peter Austin, Statistics in Medicine. Stat Med 2008;27:2055-2061; discussion 2066-2069 https://doi.org/10.1002/sim.3245
  41. Rubin DB. Using multivariate matched sampling and regression adjustment to control bias in observational studies. J Am Stat Assoc 1979;74:318-328
  42. Austin PC, Grootendorst P, Anderson GM. A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study. Stat Med 2007;26:734-753 https://doi.org/10.1002/sim.2580

Cited by

  1. Comparison of overall survival in patients with unresectable hepatic metastases with or without transarterial chemoembolization: A Propensity Score Matching Study vol.6, pp.None, 2015, https://doi.org/10.1038/srep35336
  2. Comparison of Core-Needle Biopsy and Fine-Needle Aspiration for Evaluating Thyroid Incidentalomas Detected by 18 F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography: A Pr vol.27, pp.10, 2017, https://doi.org/10.1089/thy.2017.0192
  3. A Single Perioperative Injection of Dexamethasone Decreases Nausea, Vomiting, and Pain after Laparoscopic Donor Nephrectomy vol.2017, pp.None, 2015, https://doi.org/10.1155/2017/3518103
  4. Intravenous Contrast-Induced Nephropathy-The Rise and Fall of a Threatening Idea vol.24, pp.3, 2015, https://doi.org/10.1053/j.ackd.2017.03.001
  5. Influence of neoadjuvant chemotherapy on resection of primary colorectal liver metastases: A propensity score analysis vol.116, pp.2, 2015, https://doi.org/10.1002/jso.24631
  6. Effect of Retrievable Stent Size on Endovascular Treatment of Acute Ischemic Stroke: A Multicenter Study vol.38, pp.8, 2015, https://doi.org/10.3174/ajnr.a5232
  7. Efficacy and safety of core-needle biopsy in initially detected thyroid nodules via propensity score analysis vol.7, pp.None, 2017, https://doi.org/10.1038/s41598-017-07924-z
  8. Perioperative blood transfusion does not affect recurrence-free and overall survivals after curative resection for intrahepatic cholangiocarcinoma: a propensity score matching analysis vol.17, pp.None, 2015, https://doi.org/10.1186/s12885-017-3745-z
  9. Patient satisfaction after thyroid RFA versus surgery for benign thyroid nodules: a telephone survey vol.35, pp.1, 2015, https://doi.org/10.1080/02656736.2018.1487590
  10. Methodologic Guide for Evaluating Clinical Performance and Effect of Artificial Intelligence Technology for Medical Diagnosis and Prediction vol.286, pp.3, 2018, https://doi.org/10.1148/radiol.2017171920
  11. Differentiation of intrahepatic cholangiocarcinoma from hepatocellular carcinoma in high-risk patients: A predictive model using contrast-enhanced ultrasound vol.24, pp.33, 2015, https://doi.org/10.3748/wjg.v24.i33.3786
  12. Analyses of clinical outcomes after severe pelvic fractures: an international study vol.3, pp.1, 2015, https://doi.org/10.1136/tsaco-2018-000238
  13. Effect of family practice contract services on the quality of primary care in Guangzhou, China: a cross-sectional study using PCAT-AE vol.8, pp.11, 2018, https://doi.org/10.1136/bmjopen-2017-021317
  14. Prognostic Value of the Tumor Size in Resectable Colorectal Cancer with Different Primary Locations: A Retrospective Study with the Propensity Score Matching vol.10, pp.2, 2015, https://doi.org/10.7150/jca.26882
  15. Ensuring continuity of patient care across the healthcare interface: Telephone follow‐up post‐hospitalization vol.85, pp.3, 2019, https://doi.org/10.1111/bcp.13839
  16. A survey of traditional Chinese medicine use among rheumatoid arthritis patients: a claims data-based cohort study vol.38, pp.5, 2015, https://doi.org/10.1007/s10067-018-04425-w
  17. Comparison of sequential and high-pitch-spiral coronary CT-angiography: image quality and radiation exposure vol.35, pp.7, 2015, https://doi.org/10.1007/s10554-019-01568-y
  18. Dose–response associations between metabolic indexes and the risk of comorbid type 2 diabetes mellitus among rheumatoid arthritis patients from Northern China: a case–control study vol.9, pp.7, 2015, https://doi.org/10.1136/bmjopen-2018-028011
  19. Comparison between M-score and LR-M in the reporting system of contrast-enhanced ultrasound LI-RADS vol.29, pp.8, 2019, https://doi.org/10.1007/s00330-018-5927-8
  20. Perinatal outcomes of singletons following vitrification versus slow-freezing of embryos: a multicenter cohort study using propensity score analysis vol.34, pp.9, 2019, https://doi.org/10.1093/humrep/dez095
  21. Racial association and pharmacotherapy in neonatal opioid withdrawal syndrome vol.39, pp.10, 2015, https://doi.org/10.1038/s41372-019-0440-8
  22. Comparison of outcome between intrauterine balloon tamponade and uterine artery embolization in the management of persistent postpartum hemorrhage: A propensity score‐matched cohort study vol.98, pp.11, 2015, https://doi.org/10.1111/aogs.13679
  23. Indirect comparison of novel Oral anticoagulants among Asians with non-Valvular atrial fibrillation in the real world setting: a network meta-analysis vol.19, pp.1, 2019, https://doi.org/10.1186/s12872-019-1165-5
  24. Role of the default mode resting-state network for cognitive functioning in malignant glioma patients following multimodal treatment vol.27, pp.None, 2020, https://doi.org/10.1016/j.nicl.2020.102287
  25. Neutrophil–Lymphocyte Ratio (NLR) for Predicting Clinical Outcomes in Patients with Coronary Artery Disease and Type 2 Diabetes Mellitus: A Propensity Score Matching Analysis vol.16, pp.None, 2020, https://doi.org/10.2147/tcrm.s244623
  26. Prognostic Value of Dual-Energy CT-Based Iodine Quantification versus Conventional CT in Acute Pulmonary Embolism: A Propensity-Match Analysis vol.21, pp.9, 2020, https://doi.org/10.3348/kjr.2019.0645
  27. Radiotherapy Versus Surgery–Which Is Better for Patients With T1-2N0M0 Glottic Laryngeal Squamous Cell Carcinoma? Individualized Survival Prediction Based on Web-Based Nomograms vol.10, pp.None, 2015, https://doi.org/10.3389/fonc.2020.01669
  28. An investigation into hemodynamically significant coronary artery lesions predictors assessed by fractional flow reserve: A propensity score matching analysis vol.7, pp.1, 2015, https://doi.org/10.14744/nci.2019.79058
  29. Self-reported depression in cancer survivors versus the general population: a population-based propensity score-matching analysis vol.29, pp.2, 2020, https://doi.org/10.1007/s11136-019-02339-x
  30. Concerns Remain Regarding Long-term Ozone Exposure and Respiratory Outcomes vol.180, pp.5, 2015, https://doi.org/10.1001/jamainternmed.2020.0574
  31. Healthcare Resource Use and Costs Associated with Opioid Initiation Among Patients with Newly Diagnosed Endometriosis with Commercial Insurance in the USA vol.37, pp.6, 2015, https://doi.org/10.1007/s12325-020-01361-7
  32. Effects of Pirfenidone on Echocardiographic Parameters of Left Ventricular Structure and Function in Patients with Idiopathic Pulmonary Fibrosis vol.5, pp.2, 2020, https://doi.org/10.2478/jim-2020-0009
  33. Goals-of-Care Consultations Are Associated with Lower Costs and Less Acute Care Use among Propensity-Matched Cohorts of African Americans and Whites with Serious Illness vol.23, pp.9, 2015, https://doi.org/10.1089/jpm.2019.0522
  34. Minimum 5-Year Outcomes of Robotic-assisted Primary Total Hip Arthroplasty With a Nested Comparison Against Manual Primary Total Hip Arthroplasty: A Propensity Score-Matched Study vol.28, pp.20, 2015, https://doi.org/10.5435/jaaos-d-19-00328
  35. The prognostic value of gender in gastric gastrointestinal stromal tumors: a propensity score matching analysis vol.11, pp.1, 2015, https://doi.org/10.1186/s13293-020-00321-8
  36. The Value of Prognostic Nutritional Index (PNI) on Newly Diagnosed Diffuse Large B-Cell Lymphoma Patients: A Multicenter Retrospective Study of HHLWG Based on Propensity Score Matched Analysis vol.14, pp.None, 2021, https://doi.org/10.2147/jir.s340822
  37. Safety and Feasibility of Video-Assisted Thoracoscopic Day Surgery and Inpatient Surgery in Patients With Non-small Cell Lung Cancer: A Single-Center Retrospective Cohort Study vol.8, pp.None, 2015, https://doi.org/10.3389/fsurg.2021.779889
  38. Proton Pump Inhibitors Were Associated With Reduced Pseudocysts in Acute Pancreatitis: A Multicenter Cohort Study vol.12, pp.None, 2021, https://doi.org/10.3389/fphar.2021.772975
  39. Diagnostic performance of core needle biopsy as a first‐line diagnostic tool for thyroid nodules according to ultrasound patterns: Comparison with fine needle aspiration using propensity score m vol.94, pp.3, 2015, https://doi.org/10.1111/cen.14321
  40. Lesion-Function Analysis from Multimodal Imaging and Normative Brain Atlases for Prediction of Cognitive Deficits in Glioma Patients vol.13, pp.10, 2015, https://doi.org/10.3390/cancers13102373
  41. Pre-transplant Dementia is Associated with Poor Survival After Hematopoietic Stem Cell Transplantation: A Nationwide Cohort Study with Propensity Score Matched Control vol.19, pp.2, 2015, https://doi.org/10.9758/cpn.2021.19.2.294
  42. Effectiveness of dry needling for upper extremity spasticity, quality of life and function in subacute phase stroke patients vol.39, pp.4, 2021, https://doi.org/10.1177/0964528420947426
  43. Global DNA hypermethylation in peripheral blood mononuclear cells and cardiovascular disease risk: a population-based propensity score-matched cohort study vol.75, pp.9, 2021, https://doi.org/10.1136/jech-2020-215382
  44. Thermal Ablation Versus Stereotactic Body Radiotherapy After Transarterial Chemoembolization for Inoperable Hepatocellular Carcinoma: A Propensity Score-Weighted Analysis vol.217, pp.3, 2021, https://doi.org/10.2214/ajr.20.24117
  45. The clinical effectiveness of establishing a proximal jejunum pouch after laparoscopic total gastrectomy: A propensity score-based analysis vol.45, pp.1, 2015, https://doi.org/10.1016/j.asjsur.2021.07.002