Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Austin Nichols <austinnichols@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Negative binomial regression with exposure and predictors correlated |

Date |
Tue, 31 Aug 2010 10:31:45 -0400 |

Paul <paul.seed@kcl.ac.uk>: Why not just include all RHS variables in logs rather than forming a ratio on the RHS? If the "exposure" is supposed to have a coef of one, that will be estimated; if the numerator and denominator of your ratio are supposed to have coefs of equal size and opposite sign, they will be estimated so. I would lean toward -xtpoisson, fe- myself. You might also look for "Granger causality" by hand, i.e. see if a higher outcome value predicts later higher treatment. I would guess that endogeneity is a pervasive problem here--what is the setting in which these practices are getting patients and making decisions about prescriptions? On Tue, Aug 31, 2010 at 7:41 AM, Seed, Paul <paul.seed@kcl.ac.uk> wrote: > Dear Statalist, > > I am struggling with a rather tricky modelling problem & would greatly appreciate any thoughts. > > I wish to know whether the introduction of a new treatment in Primary Care can be linked to a fall in Hospital admissions. However, all my data is at the level of the practice, not the patient or patient group. I therefore use negative binomial regression, with the number of patients as the exposure. > > The main predictor is the rate of prescribing, estimated as the total cost of prescriptions for the drug of interest, divided by number of patients in each practice. After correcting for age, gender & other drugs used, I find a strong paradoxical effect of more prescribing associated with more hospital admissions. If correct, it would appear the drug is doing harm! > > However, the number of patients with the condition per practice appears twice in the model (as divisor and as exposure), so the effect may be an artefact. > > As a sensitivity analysis, I can use a variety of different exposures and divisors: > E_diag - the number of patients with the diagnosis recorded > E_50y - The total number of patients over 50 (the condition is rarely seen below this age) > E_Pred - The predicted number of patients affected, based on the age & gender profile of the practice (typically 10 patients affected for 1 diagnosed). > > This gives 9 possible combinations of exposure and divisor. When the exposure and the divisor are both the same do I get the significant result. But I also get a significant result when using E_50y and E_pred together. ( A total of 5 results significant out of 9) > > One further complication: I actually have data repeated for 3 years. The results above generally hold when looking at one year at a time. When I combine the data & use -xtpoisson, fe-, instead of -nbreg-, only one comparison (matching E_diag with E_diag) remains significant. > > 2 questions (for those of you have read this far) > * Is there a better model to use than -nbreg- or -xtpoisson, fe-? (NOTE: xtnbreg does not generally converge, but when it does the answers are similar) > * Is it safe to ignore the anomalous result and conclude that there is no evidence of an effect ? > > I can of course supply code and output if required, but I think I have taken up enough bandwidth. > > Best Wishes, > > > Paul Seed > * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**st: Negative binomial regression with exposure and predictors correlated***From:*"Seed, Paul" <paul.seed@kcl.ac.uk>

- Prev by Date:
**st: Test for statistical significance** - Next by Date:
**st:Command for clustering without sampling** - Previous by thread:
**st: RE: R: Negative binomial regression with exposure and predictors correlated** - Next by thread:
**st: estat phtest, detail --> different results using Stata10 and Stata11 if using robust option** - Index(es):