Risk-adjusted hospital mortality rates for stroke: evidence from the Australian Stroke Clinical Registry (AuSCR)

Dominique A Cadilhac, Monique F Kilkenny, Christopher R Levi, Natasha A Lannin, Amanda G Thrift, Joosup Kim, Brenda Grabsch, Leonid Churilov, Helen M Dewey, Kelvin Hill, Steven G Faux, Rohan Grimley, Helen Castley, Peter J Hand, Andrew Wong, Geoffrey K Herkes, Melissa Gill, Douglas Crompton, Sandy Middleton, Geoffrey A Donnan and Craig S Anderson
Med J Aust 2017; 206 (8): 345-350. || doi: 10.5694/mja16.00525


Objectives: Hospital data used to assess regional variability in disease management and outcomes, including mortality, lack information on disease severity. We describe variance between hospitals in 30-day risk-adjusted mortality rates (RAMRs) for stroke, comparing models that include or exclude stroke severity as a covariate.

Design: Cohort design linking Australian Stroke Clinical Registry data with national death registrations. Multivariable models using recommended statistical methods for calculating 30-day RAMRs for hospitals, adjusted for demographic factors, ability to walk on admission, stroke type, and stroke recurrence.

Setting: Australian hospitals providing at least 200 episodes of acute stroke care, 2009–2014.

Main outcome measures: Hospital RAMRs estimated by different models. Changes in hospital rank order and funnel plots were used to explore variation in hospital-specific 30-day RAMRs; that is, RAMRs more than three standard deviations from the mean.

Results: In the 28 hospitals reporting at least 200 episodes of care, there were 16 218 episodes (15 951 patients; median age, 77 years; women, 46%; ischaemic strokes, 79%). RAMRs from models not including stroke severity as a variable ranged between 8% and 20%; RAMRs from models with the best fit, which included ability to walk and stroke recurrence as variables, ranged between 9% and 21%. The rank order of hospitals changed according to the covariates included in the models, particularly for those hospitals with the highest RAMRs. Funnel plots identified significant deviation from the mean overall RAMR for two hospitals, including one with borderline excess mortality.

Conclusions: Hospital stroke mortality rates and hospital performance ranking may vary widely according to the covariates included in the statistical analysis.

Please login with your free MJA account to view this article in full

  • Dominique A Cadilhac1,2
  • Monique F Kilkenny1,2
  • Christopher R Levi3
  • Natasha A Lannin4
  • Amanda G Thrift1
  • Joosup Kim1,2
  • Brenda Grabsch2
  • Leonid Churilov2
  • Helen M Dewey5
  • Kelvin Hill6
  • Steven G Faux7,8
  • Rohan Grimley8
  • Helen Castley9
  • Peter J Hand10
  • Andrew Wong11,12
  • Geoffrey K Herkes13
  • Melissa Gill14
  • Douglas Crompton15
  • Sandy Middleton16,17
  • Geoffrey A Donnan2
  • Craig S Anderson18,19,20

  • 1 Monash University, Melbourne, VIC
  • 2 Florey Institute of Neuroscience and Mental Health, Melbourne, VIC
  • 3 John Hunter Hospital Campus, Newcastle, NSW
  • 4 La Trobe University, Melbourne, VIC
  • 5 Eastern Health Clinical School, Monash University, Melbourne, VIC
  • 6 Stroke Foundation, Melbourne, VIC
  • 7 St Vincent's Hospital, Sydney, NSW
  • 8 Sunshine Coast Clinical School, University of Queensland, Birtinya, QLD
  • 9 Royal Hobart Hospital, Hobart, TAS
  • 10 Royal Melbourne Hospital, Melbourne, VIC
  • 11 Royal Brisbane and Women's Hospital, Brisbane, QLD
  • 12 University of Queensland, Brisbane, QLD
  • 13 Royal North Shore Hospital, Sydney, NSW
  • 14 Armidale Rural Referral Hospital, Hunter New England Local Health District, Armidale, NSW
  • 15 Northern Health, Melbourne, VIC
  • 16 St Vincent's Health Australia (Sydney), Sydney, NSW
  • 17 Australian Catholic University, Sydney, NSW
  • 18 The George Institute for Global Health, Sydney, NSW
  • 19 Royal Prince Alfred Hospital, Sydney, NSW
  • 20 The George Institute China at Peking University Health Science Center, Beijing, China


We acknowledge Joyce Lim and other staff from the George Institute for Global Health who contributed to establishing AuSCR. We also acknowledge staff from the Florey Institute of Neuroscience and Mental Health, including Karen Moss, Enna Salama, Kate Paice, Kasey Wallis, Adele Gibbs and Alison Dias from AuSCR Office, who contributed to AuSCR operations during this study period. Staff from the Stroke Foundation are acknowledged for their contributions to patient follow-up. We also thank the hospital staff for their diligence in collecting data for AuSCR, and hospital site investigators who provided data between 2010 and 2014. Dominique Cadilhac was supported by a fellowship from the National Health and Medical Research Council (NHMRC; 1063761; co-funded by the Heart Foundation), Amanda Thrift by an NHMRC fellowship (1042600), and Monique Kilkenny by an NHMRC Early Career Fellowship (1109426); Christopher Levi holds an NHMRC Practitioner Fellowship (1043913) and Craig Anderson an NHMRC Senior Principal Research Fellowship (1081356). AuSCR was supported by grants from the NHMRC (1034415), Allergan, Ipsen, Boehringer–Ingelheim, Monash University, Queensland Health, and the Stroke Foundation.

Competing interests:

No relevant disclosures.

  • 1. National Stroke Foundation. Clinical guidelines for stroke management 2010. Melbourne: Stroke Foundation, 2010. (accessed Sept 2016).
  • 2. Katzan IL, Spertus J, Bettger JP, et al. Risk adjustment of ischemic stroke outcomes for comparing hospital performance: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014; 45: 918-944.
  • 3. Scott I, Youlden D, Coory M. Are diagnosis specific outcome indicators based on administrative data useful in assessing quality of hospital care? Qual Saf Health Care 2004; 13: 32-39.
  • 4. Saposnik G, Hill MD, O’Donnell M, et al. Variables associated with 7-day, 30-day, and 1-year fatality after ischemic stroke. Stroke 2008; 39: 2318-2324.
  • 5. Hankey GJ. Long-term outcome after ischaemic stroke/transient ischaemic attack. Cerebrovasc Dis 2003; 16 Suppl 1: 14-19.
  • 6. Bureau of Health Information. Healthcare in focus 2012. How well does NSW perform? Looking out and looking in. Annual performance report: December 2012. Sydney: BHI, 2012. (accessed Sept 2016).
  • 7. National Health Performance Authority. Towards public reporting of standardised hospital mortality in Australia: Progress report. Feb 2016. (accessed Sept 2016).
  • 8. Cadilhac DA, Lannin NA, Anderson CS, et al. Protocol and pilot data for establishing the Australian Stroke Clinical Registry. Int J Stroke 2010; 5: 217-226.
  • 9. Counsell C, Dennis M, McDowall M, Warlow C. Predicting outcome after acute and subacute stroke: development and validation of new prognostic models. Stroke 2002; 33: 1041-1047.
  • 10. Cadilhac D, Kilkenny M, Churilov L, et al. Identification of a reliable subset of process indicators for clinical audit in stroke care: an example from Australia. Clinical Audit 2010; 2: 67-77.
  • 11. Katzenellenbogen JM, Vos T, Somerford P, et al. Burden of stroke in indigenous Western Australians: a study using data linkage. Stroke 2011; 42: 1515-1521.
  • 12. Dassanayake J, Gurrin L, Payne WR, et al. Is country of birth a risk factor for acute hospitalization for cardiovascular disease in Victoria, Australia? Asia Pac J Public Health 2011; 23: 280-287.
  • 13. Kapral MK, Wang H, Mamdani M, Tu JV. Effect of socioeconomic status on treatment and mortality after stroke. Stroke 2002; 33: 268-273.
  • 14. Australian Bureau of Statistics. 2033.0.55.001. Census of population and housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2011. (accessed Sept 2016).
  • 15. Hardie K, Hankey GJ, Jamrozik K, et al. Ten-year risk of first recurrent stroke and disability after first-ever stroke in the Perth Community Stroke Study. Stroke 2004; 35: 731-735.
  • 16. Ali SF, Singhal AB, Viswanathan A, et al. Characteristics and outcomes among patients transferred to a regional comprehensive stroke center for tertiary care. Stroke 2013; 44: 3148-3153.
  • 17. Fonarow GC, Smith EE, Reeves MJ, et al. Hospital-level variation in mortality and rehospitalization for Medicare beneficiaries with acute ischemic stroke. Stroke 2011; 42: 159-166.
  • 18. Feigin VL, Lawes CM, Bennett DA, et al. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 2009; 8: 355-369.
  • 19. Spiegelhalter DJ. Funnel plots for comparing institutional performance. Stat Med 2005; 24: 1185-1202.
  • 20. McCormick N, Bhole V, Lacaille D, Avina-Zubieta JA. Validity of diagnostic codes for acute stroke in administrative databases: a systematic review. PLoS One 2015; 10: e0135834.
  • 21. Appelros P, Jonsson F, Asberg S, et al. Trends in stroke treatment and outcome between 1995 and 2010: observations from Riks-Stroke, the Swedish stroke register. Cerebrovasc Dis 2014; 37: 22-29.
  • 22. Preen DB, Holman CDAJ, Lawrence DM, et al. Hospital chart review provided more accurate comorbidity information than data from a general practitioner survey or an administrative database. J Clin Epidemiol 2004; 57: 1295-1304.
  • 23. Fonarow GC, Pan W, Saver JL, et al. Comparison of 30-day mortality models for profiling hospital performance in acute ischemic stroke with vs without adjustment for stroke severity. JAMA 2012; 308: 257-264.
  • 24. Sim J, Teece L, Dennis MS, et al. Validation and recalibration of two multivariable prognostic models for survival and independence in acute stroke. PLoS One 2016; 11: e0153527.
  • 25. Xian Y, Holloway RG, Pan W, Peterson ED. Challenges in assessing hospital-level stroke mortality as a quality measure: comparison of ischemic, intracerebral hemorrhage, and total stroke mortality rates. Stroke 2012; 43: 1687-1690.


remove_circle_outline Delete Author
add_circle_outline Add Author

Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

access_time 08:41, 5 May 2017
Velandai Srikanth

We congratulate Cadilhac et al., on the important issue dealt with in their paper. We refer the authors to our paper published last year on the prediction of mortality in an international stroke trial dataset (VISTA) (1). We showed that the adjustment for stroke severity (NIHSS) at 24 hours offers the best prediction of stroke-related mortality, clearly superior to the adjustment for stroke severity at baseline. Moreover, the contribution of stroke severity was over and above that provided by the use of a comorbidity index, which the authors were unable to do in this study. The authors may also be able to replicate our analysis of comparing model performance by using methods such as net reclassification index and integrated discrimination improvement (2). We have recently also extended this modelling approach to predicting disability after ischaemic stroke and found similar results (3).
1. Phan TG, Clissold B, Ly J, Ma H, Moran C, Srikanth V, for the VISTA collaboration. Stroke severity and comorbidity index for prediction of mortality after ischemic stroke from the virtual international stroke trials archive-acute collaboration. Journal of Stroke and Cerebrovascular diseases : the official journal of National Stroke Association. 2016;25:835-842
2. Pencina MJ, D'Agostino RB, Sr., D'Agostino RB, Jr., Vasan RS. Evaluating the added predictive ability of a new marker: From area under the roc curve to reclassification and beyond. Stat Med. 2008;27:157-172; discussion 207-112
3. Phan TG, Clissold B, Ly J, Ma H, Srikanth V, for the Vista Collaboration. Predicting disability after ischemic stroke based on comorbidity index and stroke severity. Frontiers Neurology. 2017, in press; (doi: 10.3389/fneur.2017.00192)

Competing Interests: No relevant disclosures

Prof Velandai Srikanth
Peninsula Clinical School, Central Clinical School, Monash University, Melbourne, Victoria

Responses are now closed for this article.