Health Care Management Science

, Volume 13, Issue 2, pp 170–181 | Cite as

Analysing the length of care episode after hip fracture: a nonparametric and a parametric Bayesian approach

Article

Abstract

Effective utilisation of limited resources is a challenge for health care providers. Accurate and relevant information extracted from the length of stay distributions is useful for management purposes. Patient care episodes can be reconstructed from the comprehensive health registers, and in this paper we develop a Bayesian approach to analyse the length of care episode after a fractured hip. We model the large scale data with a flexible nonparametric multilayer perceptron network and with a parametric Weibull mixture model. To assess the performances of the models, we estimate expected utilities using predictive density as a utility measure. Since the model parameters cannot be directly compared, we focus on observables, and estimate the relevances of patient explanatory variables in predicting the length of stay. To demonstrate how the use of the nonparametric flexible model is advantageous for this complex health care data, we also study joint effects of variables in predictions, and visualise nonlinearities and interactions found in the data.

Keywords

Length of stay Hip fracture Bayesian analysis Multilayer perceptron Weibull mixture Covariate relevance 

References

  1. 1.
    OECD (2002) Measuring up. Improving health system performance in OECD countries. Organisation for Economic Co-operation and Development, ParisGoogle Scholar
  2. 2.
    Aday LA, Begley CE, Lairson DR, Slater CH (1998) Evaluating the healthcare system: effectiveness, efficiency, and equity, 2nd edn. Health Administration Press, ChigacoGoogle Scholar
  3. 3.
    Perleth M, Jakubowski E, Busse R (2001) What is ’best practice’ in health care? State of the art and perspectives in improving the effectiveness and efficiency of the European health care systems. Health Policy 56:235–250CrossRefGoogle Scholar
  4. 4.
    Eddy DM (1998) Performance measurement: problems and solutions (with discussion). Health Aff 17:7–41CrossRefGoogle Scholar
  5. 5.
    Loeb JM (2004) The current state of performance measurement in health care. Int J Qual Health Care 16(Suppl 1):i5–9CrossRefGoogle Scholar
  6. 6.
    Freeman T (2002) Using performance indicators to improve health care quality in the public sector: a review of the literature. Health Serv Manag Res 15:126–137CrossRefGoogle Scholar
  7. 7.
    Bird SM, Cox D, Farewell VT, Goldstein H, Holt T, Smith PC (2005) Performance indicators: good, bad, and ugly. J R Stat Soc, A 168(1):1–27CrossRefGoogle Scholar
  8. 8.
    Marazzi A, Paccaud F, Ruffieux C, Beguin, C (1998) Fitting the distributions of length of stay by parametric models. Med Care 36(6):915–927CrossRefGoogle Scholar
  9. 9.
    Fisher WH, Altaffer FB (1992) Inpatient length of stay measures: statistical and conceptual issues. Adm Policy Ment Health 19:311–320CrossRefGoogle Scholar
  10. 10.
    Marshall AH, McClean SI (2004) Using Coxian phase-type distributions to identify patient characteristics for duration of stay in hospital. Health Care Manag Sci 7:285–289CrossRefGoogle Scholar
  11. 11.
    Atienza N, García-Heras J, Muñoz-Pichardo JM, Villa R (2008) An application of mixture distributions in modelization of length of hospital stay. Stat Med 27(9):1403–1420CrossRefGoogle Scholar
  12. 12.
    Faddy MJ, McClean SI (1999) Analysing data on lengths of stay of hospital patients using phase-type distributions. Appl Stoch Models Bus Ind 15:311–317CrossRefGoogle Scholar
  13. 13.
    Lowell W, Davis G, Lajousky W, Stieffel S, Davis G, Breau M, Harvey S, Shirazi H (1997) A field trial using artificial neural networks to predict psychiatric in-patient length-of-stay. Neural Comput Appl 5:184–193CrossRefGoogle Scholar
  14. 14.
    Walczak S, Scorpio RJ, Pofahl WE (1998) Predicting hospital length of stay with neural networks. In: Proceedings of the eleventh international FLAIRS conference, pp 333–337Google Scholar
  15. 15.
    Gelman A, Carlin JB, Stern HS, Rubin DB (2003) Bayesian data analysis, 2nd edn. Chapman & Hall/CRC, Boca RatonGoogle Scholar
  16. 16.
    Bernardo JM, Smith AFM (1994) Bayesian theory. Wiley, New YorkCrossRefGoogle Scholar
  17. 17.
    Marks R, Allegrante JP, MacKenzie CR, Lane JM (2003) Hip fractures among the elderly: causes, consequences and control. Ageing Research Reviews 2(1):57–93CrossRefGoogle Scholar
  18. 18.
    Zuckerman JD (1996) Hip fracture. N Engl J Med 334(23):1519–1525CrossRefGoogle Scholar
  19. 19.
    Nurmi I, Narinen A, Lüthje P, Tanninen S (2003) Cost analysis of hip fracture treatment among the elderly for the public health services: a 1-year prospective study in 106 consecutive patients. Arch Orthop Trauma Surg 123:551–554CrossRefGoogle Scholar
  20. 20.
    Parker MJ, Todd CJ, Palmer CR, Camilleri-Ferrante C, Freeman CJ, Laxton CE, Payne BV, Rushton N (1998) Inter-hospital variations in length of hospital stay following hip fracture. Age Ageing 27:333–337CrossRefGoogle Scholar
  21. 21.
    SIGN (2002) Prevention and management of hip fracture in older people. Scottish Intercollegiate Guidelines Network, EdinburghGoogle Scholar
  22. 22.
    Sund R (2008) Methodological perspectives for register-based health system performance assessment: developing a hip fracture monitoring system in Finland. Research report 174, National Research and Development Centre for Welfare and Health. http://urn.fi/URN:ISBN:978-951-33-2132-1
  23. 23.
    Sund R, Nurmi-Lüthje I, Lüthje P, Tanninen S, Narinen A, Keskimäki I (2007) Comparing properties of audit data and routinely collected register data in case of performance assessment of hip fracture treatment in Finland. Methods Inf Med 46(5):558–566Google Scholar
  24. 24.
    Hornbrook MC, Hurtado AV, Johnson RE (1985) Health care episodes: definition, measurement and use. Med Care Res Rev 42:163–218CrossRefGoogle Scholar
  25. 25.
    Heikkinen T, Jalovaara P (2005) Four or twelve months’ follow-up in the evaluation of functional outcome after hip fracture surgery? Scand J Surg 94:59–66Google Scholar
  26. 26.
    Sund R, Liski A (2005) Quality effects of operative delay on mortality in hip fracture treatment. Qual Saf Health Care 14:371–377CrossRefGoogle Scholar
  27. 27.
    Neal RM (1996) Bayesian learning for neural networks. Springer, New YorkGoogle Scholar
  28. 28.
    Lampinen J, Vehtari A (2001) Bayesian approach for neural networks—review and case studies. Neural Netw 14(3):7–24CrossRefGoogle Scholar
  29. 29.
    Duane S, Kennedy AD, Pendleton BJ, Roweth D (1987) Hybrid Monte Carlo. Phys Lett, B 195(2):216–222CrossRefGoogle Scholar
  30. 30.
    COSY (2008) MCMC Methods for MLP and GP and Stuff (for Matlab) V2.1. http://www.lce.hut.fi/research/mm/mcmcstuff/
  31. 31.
    Ibrahim JG, Chen MH, Sinha D (2001) Bayesian survival analysis. Springer, New YorkGoogle Scholar
  32. 32.
    Diebolt J, Robert CP (1994) Estimation of finite mixture distributions through Bayesian sampling. J R Stat Soc, B 56(2):363–375Google Scholar
  33. 33.
    Neal RM (2003) Slice sampling. Ann Stat 31:705–767CrossRefGoogle Scholar
  34. 34.
    Vehtari A, Lampinen J (2002) Bayesian model assessment and comparison using cross-validation predictive densities. Neural Comput 14(10):2439–2468CrossRefGoogle Scholar
  35. 35.
    Gelman A, Pardoe I (2007) Average predictive comparisons for models with nonlinearity, interactions, and variance components. Sociol Method 37(1):23–51CrossRefGoogle Scholar
  36. 36.
    Solon JA, Feeney JJ, Jones SH, Rigg RD, Sheps CG (1969) Delineating episodes of medical care. Am J Public Health 57:401–408CrossRefGoogle Scholar
  37. 37.
    Rosen AK, Mayer-Oakes A (1999) Episodes of care: theoretical frameworks versus current operational realities. Joint Comm J Qual Improve 25:111–128Google Scholar
  38. 38.
    Gissler M, Haukka J (2004) Finnish health and social welfare registers in epidemiological research. Norsk Epidemiol 14:113–120Google Scholar
  39. 39.
    Powell AE, Davies HTO, Thomson RG (2003) Using routine comparative data to assess the quality of health care: understanding and avoiding common pitfalls. Qual Saf Health Care 12:122–128CrossRefGoogle Scholar
  40. 40.
    Ruffieux C, Marazzi A, Paccaud F (1993) Exploring models for the length of stay distribution. Soz Praventivmed 38:77–82CrossRefGoogle Scholar
  41. 41.
    Quantin C, Sauleau E, Bolard P, Mousson C, Kerkri M, Brunet Lecomte P, Moreau T, Dusserre L (1999) Modeling of high-cost patient distribution within renal failure diagnostic related group. J Clin Epidemiol 52:251–258CrossRefGoogle Scholar
  42. 42.
    Lee AH, Ng ASK, Yau KKW (2001) Determinants of maternity length of stay: a gamma mixture risk-adjusted model. Health Care Manag Sci 4:249–255CrossRefGoogle Scholar
  43. 43.
    Marshall A, Shaw B (2008) Modeling survival of hip fracture patients using a conditional phase-type distribution. In: 21st IEEE international symposium on computer-based medical systems, vol 21, pp 518–523Google Scholar
  44. 44.
    Marshall AH, McClean SI, Shapcott CM, Hastie IR, Millard PH (2001) Developing a Bayesian belief network for the management of geriatric hospital care. Health Care Manag Sci 4:25–30CrossRefGoogle Scholar
  45. 45.
    Harper PR (2002) A framework for operational modelling of hospital resources. Health Care Manag Sci 5:165–173CrossRefGoogle Scholar
  46. 46.
    Bishop CM (2006) Pattern recognition and machine learning. Springer, New YorkCrossRefGoogle Scholar
  47. 47.
    Neal RM (2006) Classification with Bayesian neural networks. Machine learning challenges. Evaluating predictive uncertainty, visual object classification, and recognising textual entailment. Lecture notes in computer science no 3944. Springer, New YorkCrossRefGoogle Scholar
  48. 48.
    Sund R, Riihimäki J, Mäkelä M, Vehtari A, Lüthje P, Huusko T, Häkkinen U (2009) Modeling the length of the care episode after hip fracture: does the type of fracture matter? Scand J Surg 98:169–174Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  1. 1.Department of Biomedical Engineering and Computational ScienceHelsinki University of Technology—TKKHelsinkiFinland
  2. 2.Service Systems Research UnitNational Institute for Health and Welfare-THLHelsinkiFinland

Personalised recommendations