Spss Poisson Regression

Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 17

Generalized Linear Models Using SPSS

Generalized Linear Models can be fitted in SPSS using the Genlin procedure. This
procedure allows you to fit models for binary outcomes, ordinal outcomes, and models
for other distributions in the exponential family (e.g., Poisson, negative binomial,
gamma).

We will be using data from Apple Tree Dental for these examples. This dataset is
based on elderly nursing home residents, and was collected as part of Grant
R03DE16976-01A1 ("Dental Utilization by Nursing Home Residents: 1986-2004",
National Institute of Dental and Craniofacial Research), Barbara J. Smith, Principal
Investigator.

There are 987 patients in this database, with baseline ages from 55 to 102 years.
They all entered the program in 1992, and were followed for a maximum of 5 follow-
up periods. Each period was from 0 days to 547 days long. A participant could have
had a period of zero days length if they came to the program, had their initial dental
visit, and then never returned for any follow-up visits. We will be taking a look at the
number of claims that these participants made for diagnostic dental services during
their first period with Apple Tree Dental, and then over the five possible periods in
the dataset. We are mainly interested in comparing three different levels of
functional dentition, FUNCTDENT, 0: Edentulous, 1: < 20 teeth, and 2: >=20 teeth. We
will also control for other covariates in the analysis.

We first import the SAS dataset, Appletree.sas7bdat, into SPSS. The commands
below were pasted, but the method is to go to File > Open > Data… and choose the file
type as SAS (*.sas7bdat, *.sd7, *.sd2, *.ssd01,*.xpt).

GET
SAS DATA='C:\Documents and Settings\kwelch\Desktop\b510\appletree.sas7bdat'.
DATASET NAME DataSet1 WINDOW=FRONT.

SAVE OUTFILE='C:\Documents and Settings\kwelch\Desktop\b510\appletree.sav'


/COMPRESSED.

We set up value labels for each level of functdent, and select for analysis only those
cases in Period 1.

value labels functdent (0) Edentulous (1) <20 teeth (2) >=20 teeth.

USE ALL.

1
COMPUTE filter_$=(Period=1).
FILTER BY filter_$.
EXECUTE.

We now take a look at the distribution of the number of Diagnostic services,


NUM_DIAGNOSTIC, using histograms paneld by FUNCTDENT.

We can see that there is a skewed distribution for this variable, which can be
expected, because these values represent counts.

Next, we modify the dataset to create a new categorical variable, NURSBEDS, which
has a value of 1: 100 or fewer beds, 2: 101-150 beds, or 3: >150 beds in the nursing
home where the participant lived. We also calculate the length of the period in years,
rather than days, so the estimated mean values for the outcome will be based on
annual, rather than daily rates of usage. We then take the natural log of the number
of years, after adding .0001 to the value, so the zero values will not be excluded. This
new variable, LOG_PERIOD_YR, will be the offset variable in our Poisson regression
model.

RECODE nbeds (MISSING=SYSMIS) (Lowest thru 100=1)


(101 thru 150=2) (151 thru Highest=3) INTO nursebeds.
EXECUTE.
Compute Period_yr = (Period_days/365.25).
Compute log_period_yr = (ln(period_yr+.0001)).
EXECUTE.

Poisson Regression Model

2
We now fit a Poisson regression model by going to Analyze > Generalized Linear
Models > Generalized Linear Models.

In the Type of Model tab, we choose Counts…Poisson loglinear. In the "Response"


tab, we choose NUM_DIAGNOSTIC as the response variable.

In the Predictors tab you can set up the categorical predictors (Factors) and the
continuous predictors (Covariates). We choose FUNCTDENT, SEX, BASEAGE, and
NURSEBEDS as "Factors" (because we wish SPSS to treat them as categorical
predictors). We choose BASEAGE as a "Covariate" because we want SPSS to treat it
as a continuous predictor (this would also be true for 0,1 indicator variables). Be sure
to choose LOG_PERIOD_YR as the OFFSET variable.

In the Model tab, we include FUNCTDENT, SEX, BASEAGE, and NURSEBEDS,


because we want to include them all as predictors in our model.

In the Estimation tab, we admire the settings and leave them as they are.

In the Statistics tab, make sure Type III is chosen as the Analysis Type, You have
the option of choosing Chi-Square Statistics as either Wald or Likelihood Ratio. We
will choose Likelihood Ratio. Click on "Include exponential parameter estimates" in the
Print section.

In the EMMEANS tab, choose FUNCTDENT as the categorical variable for which we
would like to have SPSS calculate the means of NUM_DIAGNOSTIC. When you have
arrowed FUNCTDENT into the box at the upper right of the window, select the
Contrast as Pairwise, to get comparisons of the number of diagnostic services for
each level of FUNCTDENT. Under Scale, choose Compute Means for Response (so
the results will be on the original scale).
When you've filled out everything, choose "Paste" to paste your commands, or "OK" to
run.

* Generalized Linear Models.


GENLIN Num_Diagnostic BY functdent Sex nursebeds (ORDER=ASCENDING) WITH BaseAge
/MODEL functdent Sex nursebeds BaseAge INTERCEPT=YES OFFSET=log_period_yr
DISTRIBUTION=POISSON LINK=LOG
/CRITERIA METHOD=FISHER(1) SCALE=1 COVB=MODEL MAXITERATIONS=100 MAXSTEPHALVING=5
PCONVERGE=1E-006(ABSOLUTE) SINGULAR=1E-012 ANALYSISTYPE=3(LR) CILEVEL=95
CITYPE=WALD LIKELIHOOD=FULL

3
/EMMEANS TABLES=functdent SCALE=ORIGINAL COMPARE=functdent CONTRAST=PAIRWISE
PADJUST=LSD
/MISSING CLASSMISSING=EXCLUDE
/PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION (EXPONENTIATED).

Model Information

Dependent Variable Num_Diagnostic

Probability Distribution Poisson

Link Function Log

Offset Variable log_period_yr

Case Processing Summary

N Percent

Included 981 99.4%

Excluded 6 .6%

Total 987 100.0%

Categorical Variable Information

N Percent

Factor functdent Edentulous 360 36.7%

<20 teeth 375 38.2%

>=20 teeth 246 25.1%

Total 981 100.0%

Sex F 721 73.5%

M 260 26.5%

Total 981 100.0%

nursebeds 1 125 12.7%

2 376 38.3%

3 480 48.9%

Total 981 100.0%

4
Continuous Variable Information

N Minimum Maximum Mean Std. Deviation

Dependent Variable Num_Diagnostic 981 1 12 2.30 1.785

Covariate BaseAge 981 55 102 82.73 9.090

Offset log_period_yr 981 -3.95 .40 -.3672 .68419

Goodness of Fitb

Value df Value/df

Deviance 1339.604 974 1.375

Scaled Deviance 1339.604 974

Pearson Chi-Square 2146.046 974 2.203

Scaled Pearson Chi-Square 2146.046 974

Log Likelihooda -1920.656

Akaike's Information
3855.313
Criterion (AIC)

Finite Sample Corrected AIC


3855.428
(AICC)

Bayesian Information
3889.533
Criterion (BIC)

Consistent AIC (CAIC) 3896.533

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset = log_period_yr

a. The full log likelihood function is displayed and used in computing information criteria.

b. Information criteria are in small-is-better form.

Omnibus Testa

Likelihood Ratio
Chi-Square df Sig.

338.681 6 .000

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset = log_period_yr

a. Compares the fitted model against the intercept-only model.

5
Tests of Model Effects

Type III

Likelihood Ratio
Source Chi-Square df Sig.

(Intercept) 12.988 1 .000

functdent 325.234 2 .000

Sex 9.031 1 .003

nursebeds .386 2 .824

BaseAge 2.476 1 .116

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge,
offset = log_period_yr

Parameter Estimates

95% Wald Confidence 95% Wald Confidence Interval


Interval Hypothesis Test for Exp(B)

Std. Wald Chi- d Exp(B


Parameter B Error Lower Upper Square f Sig. ) Lower Upper

(Intercept) . .
.980 .590 1.369 24.313 1 2.664 1.805 3.933
1987 000

[functdent=0] . .
-.678 -.796 -.561 128.532 1 .507 .451 .571
0598 000

[functdent=1] . .
.209 .107 .310 16.180 1 1.232 1.113 1.364
0519 000

[functdent=2] 0a . . . . . . 1 . .

[Sex=F] . .
-.148 -.244 -.053 9.212 1 .862 .783 .949
0488 002

[Sex=M] 0a . . . . . . 1 . .

[nursebeds=1.00 . .
-.042 -.174 .091 .382 1 .959 .840 1.095
] 0676 536

[nursebeds=2.00 . .
-.007 -.097 .082 .027 1 .993 .908 1.086
] 0457 870

[nursebeds=3.00
0a . . . . . . 1 . .
]

6
Parameter Estimates

BaseAge . .
.004 .000 .009 2.460 1 1.004 .999 1.009
0024 117

(Scale) 1b

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset =
log_period_yr
a. Set to zero because this parameter is redundant.
b. Fixed at the displayed value.

Estimated Marginal Means: functdent


Estimates

95% Wald Confidence Interval

functdent Mean Std. Error Lower Upper

Edentulous 1.70 .077 1.55 1.85

<20 teeth 4.12 .141 3.84 4.39

>=20 teeth 3.34 .156 3.04 3.65

Covariates appearing in the model are fixed at the following values: BaseAge=82.73

Pairwise Comparisons

95% Wald Confidence Interval for


Difference
Mean Difference
(I) functdent (J) functdent (I-J) Std. Error df Sig. Lower Upper

Edentulous <20 teeth -2.42a .149 1 .000 -2.71 -2.13

>=20 teeth -1.65a .163 1 .000 -1.97 -1.33

<20 teeth Edentulous 2.42a .149 1 .000 2.13 2.71

>=20 teeth .78a .188 1 .000 .41 1.14

>=20 teeth Edentulous 1.65a .163 1 .000 1.33 1.97

<20 teeth -.78a .188 1 .000 -1.14 -.41

Pairwise comparisons of estimated marginal means based on the original scale of dependent variable Num_Diagnostic

a. The mean difference is significant at the .05 level.

7
Overall Test Results

Wald Chi-Square df Sig.

298.515 2 .000

The Wald chi-square tests the effect of


functdent. This test is based on the linearly
independent pairwise comparisons among the
estimated marginal means.

The estimated annual number of diagnostic services for those participants who are
edentulous is 1.7, while it is 4.12 for those with < 20 teeth, and 3.34 for those with
>=20 teeth. There is a significant difference in the annual number of diagnostic
services required in Period 1 between each of the levels of functional dentition, after
controlling for the other covariates in the model.

Overdispersed Poisson Model

The value of the deviance divided by its degrees of freedom and the Pearson chi-
square divided by its degress of freedom, 1.38 and 2.20, respectively, suggest that
there might be some overdispersion. We will next fit an overdispersed Poisson model,
using Proc Genmod. To do this, we can calculate the scale as the inverse of either the
Deviance/df or Pearson Chi-square/df. We choose to use the value of Pearson Chi-
square/df. To do this, we compute a new variable in our dataset:

Compute pscale=1/2.2033.
EXECUTE.

And now, we refit our model, using PSCALE as our scale weight. To do this, go to the
Response tab, and under Scale Weight, choose PSCALE as the Scale Weight Variable.
When the revised commands are now pasted, we see the following Model subcommand:

/MODEL functdent Sex nursebeds BaseAge INTERCEPT=YES OFFSET=log_period_yr SCALEWEIGHT=pscale


DISTRIBUTION=POISSON LINK=LOG

When the Scaleweight option is specified, the model estimates are first obtained by
setting the scale to 1.0, as for the Poisson distribution; thus the parameter estimates
are unchanged from the Poisson model. Then, the scale parameter is estimated by
either the square root of the Pearson chi-square/df or the square root of the
deviance chi-square/df. The standard errors and other statistics are adjusted

8
accordingly. For example, the standard errors of the parameter estimates are
divided by the new scale statistic, making the statistical tests more conservative.

The parameter estimates and EMMEANS output from this new model is shown below.
Note that the parameter estimates are the same as in the previous model fit, but the
standard errors have been increased, resulting in more conservative statistical tests.

9
Parameter Estimates

95% Wald Confidence 95% Wald Confidence


Interval Hypothesis Test Interval for Exp(B)

Std. Wald Chi-


Parameter B Error Lower Upper Square df Sig. Exp(B) Lower Upper

(Intercept) .980 .2950 .402 1.558 11.035 1 .001 2.664 1.494 4.750

[functdent=0] -.678 .0888 -.852 -.504 58.336 1 .000 .507 .426 .604

[functdent=1] .209 .0770 .058 .360 7.343 1 .007 1.232 1.059 1.433

[functdent=2] 0a . . . . . . 1 . .

[Sex=F] -.148 .0725 -.290 -.006 4.181 1 .041 .862 .748 .994

[Sex=M] 0a . . . . . . 1 . .

[nursebeds=1.00
-.042 .1003 -.238 .155 .173 1 .677 .959 .788 1.167
]

[nursebeds=2.00
-.007 .0678 -.140 .125 .012 1 .912 .993 .869 1.134
]

[nursebeds=3.00
0a . . . . . . 1 . .
]

BaseAge .004 .0036 -.003 .011 1.116 1 .291 1.004 .997 1.011

(Scale) 1b

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset =
log_period_yr

a. Set to zero because this parameter is


redundant.

b. Fixed at the displayed value.

10
Estimates

95% Wald Confidence Interval

functdent Mean Std. Error Lower Upper

Edentulous 1.70 .114 1.47 1.92

<20 teeth 4.12 .209 3.71 4.53

>=20 teeth 3.34 .231 2.89 3.80

Covariates appearing in the model are fixed at the following values:


BaseAge=82.73

Pairwise Comparisons

95% Wald Confidence Interval for


Difference
Mean Difference
(I) functdent (J) functdent (I-J) Std. Error df Sig. Lower Upper

Edentulous <20 teeth -2.42a .221 1 .000 -2.86 -1.99

>=20 teeth -1.65a .242 1 .000 -2.12 -1.17

<20 teeth Edentulous 2.42a .221 1 .000 1.99 2.86

>=20 teeth .78a .278 1 .005 .23 1.32

>=20 teeth Edentulous 1.65a .242 1 .000 1.17 2.12

<20 teeth -.78a .278 1 .005 -1.32 -.23

Pairwise comparisons of estimated marginal means based on the original scale of dependent variable Num_Diagnostic

a. The mean difference is significant at the .05 level.

Overall Test Results

Wald Chi-Square df Sig.

135.485 2 .000

The Wald chi-square tests the effect of


functdent. This test is based on the linearly
independent pairwise comparisons among the
estimated marginal means.

Negative Binomial Model

11
We now refit the model, using a Negative Binomial distribution for the response
variable, to do this, go to the Type of Model tab, and select Negative Binomial with
Log ling. Be sure you go to the Response tab and remove the Scale Weight Variable.
The pasted resulting portion of the commands for the Model subcommand are shown
below:

/MODEL functdent Sex nursebeds BaseAge INTERCEPT=YES OFFSET=log_period_yr


DISTRIBUTION=NEGBIN(1) LINK=LOG

***You need to change this syntax, as shown below (i.e., replace the (1) with (MLE)
after the NEGBIN portion of the syntax) to get the correct Maximum Likelihood
estimate of the Negative Binomial Dispersion Parameter:

/MODEL functdent Sex nursebeds BaseAge INTERCEPT=YES OFFSET=log_period_yr


DISTRIBUTION=NEGBIN(MLE) LINK=LOG

Selected portions from the output from this Negative Binomial regression model fit
are shown below. Note that the deviance/df and Pearson chi-square/df are now closer
to 1.0, so this is an improvement over the original Poisson Model.

Model Information

Dependent Variable Num_Diagnostic

Probability Distribution Negative binomial (1)

Link Function Log

Offset Variable log_period_yr

12
Goodness of Fitb

Value df Value/df

Deviance 1010.314 973 1.038

Scaled Deviance 1010.314 973

Pearson Chi-Square 1715.272 973 1.763

Scaled Pearson Chi-Square 1715.272 973

Log Likelihooda -1892.110

Akaike's Information
3800.220
Criterion (AIC)

Finite Sample Corrected AIC


3800.368
(AICC)

Bayesian Information
3839.328
Criterion (BIC)

Consistent AIC (CAIC) 3847.328

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset = log_period_yr

a. The full log likelihood function is displayed and used in computing information
criteria.

b. Information criteria are in small-is-better form.

Omnibus Testa

Likelihood Ratio
Chi-Square df Sig.

236.106 6 .000

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset = log_period_yr

a. Compares the fitted model against the intercept-only model.

13
Tests of Model Effects

Type III

Likelihood Ratio
Source Chi-Square df Sig.

(Intercept) 9.888 1 .002

functdent 226.222 2 .000

Sex 6.347 1 .012

nursebeds .552 2 .759

BaseAge 1.857 1 .173

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset = log_period_yr

14
Parameter Estimates

95% Wald Confidence 95% Wald Confidence


Interval Hypothesis Test Interval for Exp(B)

Std. Wald Chi-


Parameter B Error Lower Upper Square df Sig. Exp(B) Lower Upper

(Intercept) 1.009 .2363 .546 1.472 18.225 1 .000 2.742 1.726 4.358

[functdent=0] -.691 .0689 -.826 -.556 100.614 1 .000 .501 .438 .574

[functdent=1] .224 .0621 .103 .346 13.054 1 .000 1.252 1.108 1.414

[functdent=2] 0a . . . . . . 1 . .

[Sex=F] -.148 .0584 -.263 -.034 6.417 1 .011 .862 .769 .967

[Sex=M] 0a . . . . . . 1 . .

[nursebeds=1.00
-.059 .0795 -.215 .097 .546 1 .460 .943 .807 1.102
]

[nursebeds=2.00
-.010 .0543 -.116 .096 .034 1 .854 .990 .890 1.101
]

[nursebeds=3.00
0a . . . . . . 1 . .
]

BaseAge .004 .0029 -.002 .010 1.853 1 .173 1.004 .998 1.010

(Scale) 1b

(Negative
.145 .0237 .105 .200
binomial)

Dependent Variable: Num_Diagnostic


Model: (Intercept), functdent, Sex, nursebeds, BaseAge, offset =
log_period_yr

a. Set to zero because this parameter is


redundant.

b. Fixed at the displayed value.

Estimated Marginal Means: functdent

15
Estimates

95% Wald Confidence Interval

functdent Mean Std. Error Lower Upper

Edentulous 1.73 .089 1.56 1.91

<20 teeth 4.32 .183 3.97 4.68

>=20 teeth 3.45 .191 3.08 3.83

Covariates appearing in the model are fixed at the following values: BaseAge=82.73

Pairwise Comparisons

95% Wald Confidence Interval for


Difference
Mean Difference
(I) functdent (J) functdent (I-J) Std. Error df Sig. Lower Upper

Edentulous <20 teeth -2.59a .188 1 .000 -2.96 -2.22

>=20 teeth -1.72a .197 1 .000 -2.11 -1.34

<20 teeth Edentulous 2.59a .188 1 .000 2.22 2.96

>=20 teeth .87a .235 1 .000 .41 1.33

>=20 teeth Edentulous 1.72a .197 1 .000 1.34 2.11

<20 teeth -.87a .235 1 .000 -1.33 -.41

Pairwise comparisons of estimated marginal means based on the original scale of dependent variable Num_Diagnostic

a. The mean difference is significant at the .05 level.

Overall Test Results

Wald Chi-Square df Sig.

218.816 2 .000

The Wald chi-square tests the effect of


functdent. This test is based on the linearly
independent pairwise comparisons among the
estimated marginal means.

16
There are some minor differences in the model estimates and standard errors for
this negative binomial model vs. the original Poisson model, but the model fit
statistics are better.

17

You might also like