Colin G Weaver1, Pietro Ravani2, Matthew J Oliver3, Peter C Austin4, Robert R Quinn2. 1. Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada. 2. Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada Department of Medicine, University of Calgary, Calgary, Alberta, Canada. 3. Department of Medicine, University of Toronto, Toronto, Ontario, Canada Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada. 4. Institute for Clinical Evaluative Sciences, University of Toronto, Toronto, Ontario, Canada Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada.
Abstract
BACKGROUND: Poisson regression is commonly used to analyze hospitalization data when outcomes are expressed as counts (e.g. number of days in hospital). However, data often violate the assumptions on which Poisson regression is based. More appropriate extensions of this model, while available, are rarely used. METHODS: We compared hospitalization data between 206 patients treated with hemodialysis (HD) and 107 treated with peritoneal dialysis (PD) using Poisson regression and compared results from standard Poisson regression with those obtained using three other approaches for modeling count data: negative binomial (NB) regression, zero-inflated Poisson (ZIP) regression and zero-inflated negative binomial (ZINB) regression. We examined the appropriateness of each model and compared the results obtained with each approach. RESULTS: During a mean 1.9 years of follow-up, 183 of 313 patients (58%) were never hospitalized (indicating an excess of 'zeros'). The data also displayed overdispersion (variance greater than mean), violating another assumption of the Poisson model. Using four criteria, we determined that the NB and ZINB models performed best. According to these two models, patients treated with HD experienced similar hospitalization rates as those receiving PD {NB rate ratio (RR): 1.04 [bootstrapped 95% confidence interval (CI): 0.49-2.20]; ZINB summary RR: 1.21 (bootstrapped 95% CI 0.60-2.46)}. Poisson and ZIP models fit the data poorly and had much larger point estimates than the NB and ZINB models [Poisson RR: 1.93 (bootstrapped 95% CI 0.88-4.23); ZIP summary RR: 1.84 (bootstrapped 95% CI 0.88-3.84)]. CONCLUSIONS: We found substantially different results when modeling hospitalization data, depending on the approach used. Our results argue strongly for a sound model selection process and improved reporting around statistical methods used for modeling count data.
BACKGROUND: Poisson regression is commonly used to analyze hospitalization data when outcomes are expressed as counts (e.g. number of days in hospital). However, data often violate the assumptions on which Poisson regression is based. More appropriate extensions of this model, while available, are rarely used. METHODS: We compared hospitalization data between 206 patients treated with hemodialysis (HD) and 107 treated with peritoneal dialysis (PD) using Poisson regression and compared results from standard Poisson regression with those obtained using three other approaches for modeling count data: negative binomial (NB) regression, zero-inflated Poisson (ZIP) regression and zero-inflated negative binomial (ZINB) regression. We examined the appropriateness of each model and compared the results obtained with each approach. RESULTS: During a mean 1.9 years of follow-up, 183 of 313 patients (58%) were never hospitalized (indicating an excess of 'zeros'). The data also displayed overdispersion (variance greater than mean), violating another assumption of the Poisson model. Using four criteria, we determined that the NB and ZINB models performed best. According to these two models, patients treated with HD experienced similar hospitalization rates as those receiving PD {NB rate ratio (RR): 1.04 [bootstrapped 95% confidence interval (CI): 0.49-2.20]; ZINB summary RR: 1.21 (bootstrapped 95% CI 0.60-2.46)}. Poisson and ZIP models fit the data poorly and had much larger point estimates than the NB and ZINB models [Poisson RR: 1.93 (bootstrapped 95% CI 0.88-4.23); ZIP summary RR: 1.84 (bootstrapped 95% CI 0.88-3.84)]. CONCLUSIONS: We found substantially different results when modeling hospitalization data, depending on the approach used. Our results argue strongly for a sound model selection process and improved reporting around statistical methods used for modeling count data.
Authors: Matthew J Oliver; Ahmed A Al-Jaishi; Stephanie N Dixon; Jeffrey Perl; Arsh K Jain; Susan D Lavoie; Danielle M Nash; J Michael Paterson; Charmaine E Lok; Robert R Quinn Journal: Clin J Am Soc Nephrol Date: 2016-07-27 Impact factor: 8.237
Authors: Ladan Golestaneh; Kerri L Cavanaugh; Yungtai Lo; Angelo Karaboyas; Michal L Melamed; Tanya S Johns; Keith C Norris Journal: Am J Kidney Dis Date: 2020-07-13 Impact factor: 8.860
Authors: David J T Campbell; Braden J Manns; Robert G Weaver; Brenda R Hemmelgarn; Kathryn M King-Shier; Claudia Sanmartin Journal: BMC Med Date: 2017-02-15 Impact factor: 8.775