count data analysis

[Raed Alzyoud]

Dear All
Anyone can help me to do the basic descriptive data analysis and then run a regression model to predict the statistically significant variable in predicting the disease?
The table shows clinical presentation for subcategories of a group of diseases called (PID: primary immunodeficiency) and the numbers represent the frequency of each clinical presentation in each subcategory of a total of 467 patients

[Dirk Enzmann]

Before posting on the Stata Forum you should have read the forum's FAQ -- your post shows that you didn't do that. In your case you should carefully read section 12 "What should I say about the commands and data I use?".

[Raed Alzyoud]

Dear Dirk
Thanks for the guidance, based on FAQ 12.2, this is my data!

* Example generated by -dataex-. For more info, type help dataex
clear
input long PIDcat byte(SCID CID synCID AB dysreg Phagocytic innate complement AIS BMF) int Total
38 89 66 42 79 65 50 22 13 40 1 467
27 77 45 23 63  8 14 10  3 14 1 258
35 39 28 21 50  7 14 10  3  9 1 182
50 33 12  5  0  5  0  0  0  5 0  60
34  6  3  0  0  0  0  0  0  0 0   9
42  3  3  0 12  1  0  0  0  0 0  19
 9  3  1  0  0  0  0  0  0  0 0   4
 4  2  3  0  0  1  0  0  0  0 0   6
12  2  3  3  4  1  2  1  0  0 0  16
30  0  3  1 36  0  2  0  4  0 0  46
19 59 32  8 20 22 25 12  5 29 1 213
47 26  8  0  0  0  0  5  0  0 0  39
15 27 10  5 12  4 10  4  3  6 1  82
18 17 15  3 10 18 19  7  0  7 1  97
49  1  3  0  1  4  4  0  0  1 0  14
48  3  2  3 12  2  3  4  1  4 0  34
20  9  3  2  5 11  6  4  2  4 0  46
44 17  5  5  3  5  4  3  2  2 0  46
52  1  2  3  0  5  2  0  0  2 0  15
51  4  3  3  0  2 10  4  0  2 0  28
17  7  1  0  0  0  0  0  0  0 0   8
32  9  0  0  0  0  0  0  0  0 0   9
26  5  3  3 12 28 12  8  3  8 0  82
53 11  4  1  2 38  8  6  1  3 0  74
45 16  2  2  1 45 12  8  1  4 0  91
 1 16  6 14 24 22 10  7 10 29 0 138
43  8  4  6 11  6  6  3  3  8 0  55
 2  3  1  2  2  5  0  0  2  3 0  18
39  1  0  4  2  3  0  2  3  1 0  16
41  1  0  1  2  5  0  0  0 14 0  23
36  0  0  0  1  1  0  0  2  1 0   5
11  0  0  0  4  0  0  0  0  0 0   4
 3  0  0  1  0  2  0  0  0  0 0   3
 2  3  1  0  2  0  4  2  0  2 0  14
 7 21 12 13 23 10  8  5  4  0 0  96
14 12 13 22 12 24  9  2  2  3 1 100
28  2  0  3  0  0  0  0  0  0 0   5
21  4  1  0  0 46  1  2  0  2 0  56
25 89 66 23 65 47  0  1  0  2 1 294
 5 63 46 21 47 49 50 22  5  4 1 308
 6 61 35 15  8 39  0  3  0  2 1 164
40 60 27  7  6 34 50 13  0  2 1 200
22 65 27  1  0 22 25  4  0  0 0 144
24 13  7  6  8 35 10  2  3 10 0  94
10  2  0  0  3  3  1  0  0  8 0  17
23  6  2  4  0  1 13  2  0  0 0  28
13  2  5  1  0  3  1  0  0  0 0  12
16  2  1  1  2 12  0  0  0  0 0  18
 8 17  3  0  0  0  8  5  0  0 0  33
29  8  5  1 10  4  5  4  3  3 0  43
37 13  5  1  9  3  8  5  1  3 1  49
33  0  0  0  0  0  2  1  0  1 0   4
31  4  0  2  0  3  9  0  0  3 0  21
46  2  1  0  1  4  0  0  0  0 0   8
end
label values PIDcat PIDcat
label def PIDcat 1 "AID", modify
label def PIDcat 2 "AIE", modify
label def PIDcat 3 "AIH", modify
label def PIDcat 4 "AP", modify
label def PIDcat 5 "Ab", modify
label def PIDcat 6 "Antiviral", modify
label def PIDcat 7 "Atopy", modify
label def PIDcat 8 "BCGosis", modify
label def PIDcat 9 "BO", modify
label def PIDcat 10 "Biologics", modify
label def PIDcat 11 "CD", modify
label def PIDcat 12 "CLD", modify
label def PIDcat 13 "CMV", modify
label def PIDcat 14 "CNS", modify
label def PIDcat 15 "Chrdiarrhea", modify
label def PIDcat 16 "EBV", modify
label def PIDcat 17 "FMGvHD", modify
label def PIDcat 18 "FTT", modify
label def PIDcat 19 "GI", modify
label def PIDcat 20 "GIinfections", modify
label def PIDcat 21 "HLH", modify
label def PIDcat 22 "HSCT", modify
label def PIDcat 23 "IFI", modify
label def PIDcat 24 "IS", modify
label def PIDcat 25 "IVIG", modify
label def PIDcat 26 "LN", modify
label def PIDcat 27 "LRT", modify
label def PIDcat 28 "Malignancy", modify
label def PIDcat 29 "Meningitis", modify
label def PIDcat 30 "OM", modify
label def PIDcat 31 "Organabscess", modify
label def PIDcat 32 "Osphenotype", modify
label def PIDcat 33 "Osteomyelitis", modify
label def PIDcat 34 "PCjP", modify
label def PIDcat 35 "Pneumonia", modify
label def PIDcat 36 "SLE", modify
label def PIDcat 37 "Septicemia", modify
label def PIDcat 38 "Totalobs", modify
label def PIDcat 39 "Vasculitis", modify
label def PIDcat 40 "antifungal", modify
label def PIDcat 41 "arthritis", modify
label def PIDcat 42 "bronchiactasis", modify
label def PIDcat 43 "cytopenias", modify
label def PIDcat 44 "dermatitis", modify
label def PIDcat 45 "hepatomegaly", modify
label def PIDcat 46 "necFasciitis", modify
label def PIDcat 47 "oralcand", modify
label def PIDcat 48 "oralulcers", modify
label def PIDcat 49 "perianald", modify
label def PIDcat 50 "pneumonitis", modify
label def PIDcat 51 "skinabs", modify
label def PIDcat 52 "skinulcers", modify
label def PIDcat 53 "splenomegaly", modify

------------------ copy up to and including the previous line ------------------

[Dirk Enzmann]

Now the question is: What do you want to describe and what is your prediction model (i.e. what do you want to predict with which variables)?

... predict the statistically significant variable ...

sounds strange to me: Only a predictor can be "statistically significant", and whether it is statistically significant depends on the other variables in your model.

[Raed Alzyoud]

For example, the most common presentation of all subcategories is LRT ( lower respiratory tract) in 258 patients out of 467. Is this statistically significant compared to other variables? like anyone presented in 30 or 50 patients?
Thanks for your patience!

[Dirk Enzmann]

I can't help you here because I have no medical expertise (e.g. I don't know what CID means and can't formulate expectations about frequencies and relationships between diagnoses and symptoms, which is actually your task). But without any hypothesis (for example a uniform distribution of the categories) you can't test anything.

If CID represents the frequencies of the categories of PID (note, however, that there are two rows with a PIDcat code "AIE"; also there is a row with a PIDcat code "totalobs"), you can obtain a simple description of PIDcat by using CID as a frequency weight:

. tabulate PIDcat [fw=CID]

        PIDcat |      Freq.     Percent        Cum.
---------------+-----------------------------------
           AID |          6        1.14        1.14
           AIE |          2        0.38        1.52
            AP |          3        0.57        2.08
            Ab |         46        8.71       10.80
     Antiviral |         35        6.63       17.42
         Atopy |         12        2.27       19.70
       BCGosis |          3        0.57       20.27
            BO |          1        0.19       20.45
           CLD |          3        0.57       21.02
           CMV |          5        0.95       21.97
           CNS |         13        2.46       24.43
   Chrdiarrhea |         10        1.89       26.33
           EBV |          1        0.19       26.52
        FMGvHD |          1        0.19       26.70
           FTT |         15        2.84       29.55
            GI |         32        6.06       35.61
  GIinfections |          3        0.57       36.17
           HLH |          1        0.19       36.36
          HSCT |         27        5.11       41.48
           IFI |          2        0.38       41.86
            IS |          7        1.33       43.18
          IVIG |         66       12.50       55.68
            LN |          3        0.57       56.25
           LRT |         45        8.52       64.77
    Meningitis |          5        0.95       65.72
            OM |          3        0.57       66.29
          PCjP |          3        0.57       66.86
     Pneumonia |         28        5.30       72.16
    Septicemia |          5        0.95       73.11
      Totalobs |         66       12.50       85.61    <-- ?
    antifungal |         27        5.11       90.72
bronchiactasis |          3        0.57       91.29
    cytopenias |          4        0.76       92.05
    dermatitis |          5        0.95       92.99
  hepatomegaly |          2        0.38       93.37
  necFasciitis |          1        0.19       93.56
      oralcand |          8        1.52       95.08
    oralulcers |          2        0.38       95.45
     perianald |          3        0.57       96.02
   pneumonitis |         12        2.27       98.30
       skinabs |          3        0.57       98.86
    skinulcers |          2        0.38       99.24
  splenomegaly |          4        0.76      100.00
---------------+-----------------------------------
         Total |        528      100.00

or you can install -fre- from SSC and create a table of weighted frequencies in descending order:

. cap which fre
. if _rc ssc install fre // install fre if necessary

. fre PIDcat [fw=CID], desc all

PIDcat
-----------------------------------------------------------------------
                          |      Freq.    Percent      Valid       Cum.
--------------------------+--------------------------------------------
Valid   25 IVIG           |         66      12.50      12.50      12.50
        38 Totalobs       |         66      12.50      12.50      25.00    <-- ?
        5  Ab             |         46       8.71       8.71      33.71
        27 LRT            |         45       8.52       8.52      42.23
        6  Antiviral      |         35       6.63       6.63      48.86
        19 GI             |         32       6.06       6.06      54.92
        35 Pneumonia      |         28       5.30       5.30      60.23
        22 HSCT           |         27       5.11       5.11      65.34
        40 antifungal     |         27       5.11       5.11      70.45
        18 FTT            |         15       2.84       2.84      73.30
        14 CNS            |         13       2.46       2.46      75.76
        7  Atopy          |         12       2.27       2.27      78.03
        50 pneumonitis    |         12       2.27       2.27      80.30
        15 Chrdiarrhea    |         10       1.89       1.89      82.20
        47 oralcand       |          8       1.52       1.52      83.71
        24 IS             |          7       1.33       1.33      85.04
        1  AID            |          6       1.14       1.14      86.17
        13 CMV            |          5       0.95       0.95      87.12
        29 Meningitis     |          5       0.95       0.95      88.07
        37 Septicemia     |          5       0.95       0.95      89.02
        44 dermatitis     |          5       0.95       0.95      89.96
        43 cytopenias     |          4       0.76       0.76      90.72
        53 splenomegaly   |          4       0.76       0.76      91.48
        4  AP             |          3       0.57       0.57      92.05
        8  BCGosis        |          3       0.57       0.57      92.61
        12 CLD            |          3       0.57       0.57      93.18
        20 GIinfections   |          3       0.57       0.57      93.75
        26 LN             |          3       0.57       0.57      94.32
        30 OM             |          3       0.57       0.57      94.89
        34 PCjP           |          3       0.57       0.57      95.45
        42 bronchiactasis |          3       0.57       0.57      96.02
        49 perianald      |          3       0.57       0.57      96.59
        51 skinabs        |          3       0.57       0.57      97.16
        2  AIE            |          2       0.38       0.38      97.54
        23 IFI            |          2       0.38       0.38      97.92
        45 hepatomegaly   |          2       0.38       0.38      98.30
        48 oralulcers     |          2       0.38       0.38      98.67
        52 skinulcers     |          2       0.38       0.38      99.05
        9  BO             |          1       0.19       0.19      99.24
        16 EBV            |          1       0.19       0.19      99.43
        17 FMGvHD         |          1       0.19       0.19      99.62
        21 HLH            |          1       0.19       0.19      99.81
        46 necFasciitis   |          1       0.19       0.19     100.00
        Total             |        528     100.00     100.00          
-----------------------------------------------------------------------

If you can state your research question more clearly, perhaps others can give you better advice.

[Raed Alzyoud]

Dear Dirk
thanks for your help
I will follow your advice and test my hypothesis
I appreciate it