Analysis of paediatric CKD registry data in India

First step is to remove the blank rows from 1 to 52

Next, we clean the state variable. Clean the states, recode missing rows and gibberish to unknown and remove non-India cases

# A tibble: 26 × 3
   state_new          n  freq
   <fct>          <int> <dbl>
 1 Unknown          360 15.5 
 2 Maharashtra      302 13.0 
 3 Tamil Nadu       259 11.2 
 4 New Delhi        251 10.8 
 5 Karnataka        232 10.0 
 6 Uttar Pradesh    188  8.12
 7 West Bengal      186  8.03
 8 Haryana          105  4.53
 9 Madhya Pradesh    92  3.97
10 Bihar             67  2.89
# ℹ 16 more rows

# A tibble: 2 × 3
  state_valid     n  freq
  <chr>       <int> <dbl>
1 invalid       360  15.5
2 valid        1956  84.5

# A tibble: 17 × 2
   State          n
   <chr>      <int>
 1 <NA>         132
 2 M S           93
 3 X             39
 4 M.S           21
 5 5000-20000    20
 6 XX            17
 7 > 20000       15
 8 < 5000         9
 9 M  S           3
10 AMP            2
11 BAMP HARE      2
12 XXX            2
13 ASRA           1
14 MS             1
15 hjkkj          1
16 jh             1
17 jkhkh          1

We can see that 15.5% of the state variable is either missing or gibberish.

Next, is the Gender variable

# A tibble: 2 × 3
  Gender     n  freq
  <chr>  <int> <dbl>
1 Male    1761  76.0
2 Female   555  24.0

We have complete information on Gender. 75% are males!!!.

Next, we clean the AgeYears, AgeMonths, AgeDays variables and combine them into a single variable age_new

# A tibble: 21 × 3
   `as.factor(floor(age_new))`     n  freq
   <fct>                       <int> <dbl>
 1 0                              70  3.02
 2 1                             195  8.42
 3 2                             132  5.70
 4 3                             137  5.92
 5 4                             138  5.96
 6 5                             155  6.69
 7 6                             143  6.17
 8 7                             165  7.12
 9 8                             174  7.51
10 9                             163  7.04
# ℹ 11 more rows


invalid   valid 
     12    2304

There is 1 case whose age was >=19 years and removed from the dataset. There are 11 cases with missing age.

Next, we clean the date of admission variable InsertionDate and create a new variable year_new

# A tibble: 10 × 3
   year_new     n    freq
      <dbl> <int>   <dbl>
 1     2008     1  0.0432
 2     2013     2  0.0864
 3     2014     2  0.0864
 4     2015     4  0.173 
 5     2016     8  0.346 
 6     2017   243 10.5   
 7     2018   433 18.7   
 8     2019   318 13.7   
 9     2020   238 10.3   
10       NA  1066 46.0


invalid   valid 
   1083    1232

There are very few cases with year < 2017 and may have to be removed from the dataset.(Decision to be taken) There are also 935 (~43%) cases with missing year information. These two are invalid dates.

Next, we clean the diagnosis/etiology variables BasicDiagnosis, Etiologynew, broadetiologycategory and create a new variable diagnosis_new

# A tibble: 5 × 3
  diagnosis_new                 n  freq
  <chr>                     <int> <dbl>
1 Other                      1502 64.9 
2 Glomerulomephritis          333 14.4 
3 Unknown                     329 14.2 
4 Polycystic Kidney Disease   124  5.36
5 CAKUT                        27  1.17

# A tibble: 11 × 3
   Etiologynew                    n   freq
   <chr>                      <int>  <dbl>
 1 Obstructive uropathy         515 34.3  
 2 Hypoplasia-dysplasia         306 20.4  
 3 Reflux nephropathy           290 19.3  
 4 Neurogenic bladder           131  8.72 
 5 Tubulointerstitial disease   112  7.46 
 6 Hemolytic uremic syndrome     41  2.73 
 7 Chronic pyelonephritis        36  2.40 
 8 Inherited tubular disease     32  2.13 
 9 Nephrolithiasis               18  1.20 
10 Renovascular disorders        12  0.799
11 AKIsequelae                    9  0.599

Etiologynew is the most complete and clean variable and hence we will use this variable for further analysis. Categorizing this variable into GBD relevant etiologies, we can see that most common is the “Other group” and we can see the distribution of the etiologies within this group. There was no case with HTN or DM as an etiology of CKD.

Next, we clean the CKD stage variable.

# A tibble: 4 × 3
  CKDstagenew     n   freq
        <dbl> <int>  <dbl>
1           5  1063 45.9  
2           4   731 31.6  
3           3   508 21.9  
4          NA    13  0.562

For the CKD stage, the CKDstagenew variable is best one to use as it has the most complete information. There are only 13 cases with missing stage information. #### Table 1 Summary of key variables in paediatric CKD registry

Characteristic	N = 2,315¹
age_new	8.0 (4.0, 11.0)
Unknown	11
age_valid
invalid	11 (0.5%)
valid	2,304 (100%)
Gender
Female	555 (24%)
Male	1,760 (76%)
state_new
Andhra Pradesh	58 (2.5%)
Assam	19 (0.8%)
Bihar	67 (2.9%)
Chhattisgarh	17 (0.7%)
Gujarat	2 (<0.1%)
Haryana	105 (4.5%)
Himachal Pradesh	1 (<0.1%)
Jammu and Kashmir	8 (0.3%)
Jharkhand	19 (0.8%)
Karnataka	232 (10%)
Kerala	51 (2.2%)
Madhya Pradesh	92 (4.0%)
Maharashtra	302 (13%)
Manipur	1 (<0.1%)
New Delhi	251 (11%)
Odisha	7 (0.3%)
Other Union Territories	12 (0.5%)
Punjab	8 (0.3%)
Rajasthan	62 (2.7%)
Tamil Nadu	259 (11%)
Telangana	3 (0.1%)
Tripura	1 (<0.1%)
Unknown	360 (16%)
Uttar Pradesh	188 (8.1%)
Uttarakhand	4 (0.2%)
West Bengal	186 (8.0%)
state_valid
invalid	360 (16%)
valid	1,955 (84%)
year_new
2008	1 (<0.1%)
2013	2 (<0.1%)
2014	2 (<0.1%)
2015	4 (0.2%)
2016	8 (0.3%)
2017	243 (10%)
2018	433 (19%)
2019	318 (14%)
2020	238 (10%)
Missing	1,066 (46%)
diagnosis_new
CAKUT	27 (1.2%)
Glomerulomephritis	333 (14%)
Other	1,502 (65%)
Polycystic Kidney Disease	124 (5.4%)
Unknown	329 (14%)
Etiologynew
AKIsequelae	9 (0.4%)
CGN	333 (14%)
Chronic pyelonephritis	36 (1.6%)
Cystic kidney disease	124 (5.4%)
Hemolytic uremic syndrome	41 (1.8%)
Hypoplasia-dysplasia	306 (13%)
Inherited tubular disease	32 (1.4%)
Nephrolithiasis	18 (0.8%)
Neurogenic bladder	131 (5.7%)
Obstructive uropathy	515 (22%)
Other CAKUT	27 (1.2%)
Reflux nephropathy	290 (13%)
Renovascular disorders	12 (0.5%)
Tubulointerstitial disease	112 (4.8%)
Undetermined	329 (14%)
CKDstagenew
3	508 (22%)
4	731 (32%)
5	1,063 (46%)
Missing	13 (0.6%)
¹ Median (IQR); n (%)

In this plot of age distribution, we can see that the majority of patients are between 1 and 12 years old. This does not look like normal distribution.

Majority of cases are in the 0-5 year age group.

As seen before, majority are male children. This points to a major selection bias.

In this plot we can see that the majority of patients are from the states of Maharashtra, Tamil Nadu, New Delhi, Karnataka. Slightly >10% have no state information.

Most cases are between 2017 and 2020.

There are more cases in Stage 5 that any other stage.There are no cases in Stage 1-2!!!

The majority of cases are in the “Other” category.

In this plot we can see the distribution of the etiology of the “Other category. Most of the etiologies are related to anatomical changes of the urinary tract.

There is no variation in the distribution of CKD stages by age group

Across all age groups, the majority of cases have “Other” as the etiology following by GN and unknown. As age increases the proportion of cases with “Other” and GN as the etiology decreases and the proportion of cases with unknown etiology increases.

There is no difference in the distribution of diagnosis by year of admission.

# A tibble: 59 × 2
   state              n
   <chr>          <int>
 1 Uttar Pradesh   7549
 2 West Bengal     7202
 3 Maharashtra     5221
 4 Tamil Nadu      4832
 5 Odisha          2768
 6 Gujarat         2729
 7 Andhra Pradesh  2622
 8 Jharkhand       2019
 9 Delhi           1715
10 Bihar           1709
# ℹ 49 more rows

# A tibble: 31 × 2
   state                 n
   <chr>             <int>
 1 Andhra Pradesh     2622
 2 Arunachal Pradesh     6
 3 Assam               930
 4 Bihar              1709
 5 Chhattisgarh        351
 6 Delhi              1715
 7 Goa                  22
 8 Gujarat            2729
 9 Haryana             221
10 Himachal Pradesh     47
# ℹ 21 more rows

# A tibble: 148 × 2
   height_clean     n
          <dbl> <int>
 1            1     1
 2            9     2
 3           15     1
 4           20     4
 5           25     3
 6           27     1
 7           30     4
 8           33     2
 9           35     2
10           40     1
# ℹ 138 more rows

# A tibble: 210 × 2
   serum_creatinine_clean     n
                    <dbl> <int>
 1                   0.3     29
 2                   0.4     49
 3                   0.5     49
 4                   0.6     58
 5                   0.7     51
 6                   0.8     68
 7                   0.84     1
 8                   0.9     63
 9                   1       58
10                   1.1     30
# ℹ 200 more rows

# A tibble: 150 × 2
   `floor(eGFR_schwartz)`     n
                    <dbl> <int>
 1                      1     2
 2                      2    12
 3                      3    39
 4                      4    69
 5                      5    80
 6                      6    92
 7                      7    84
 8                      8    72
 9                      9    64
10                     10    48
# ℹ 140 more rows

# A tibble: 5 × 2
  CKD_stage     n
  <chr>     <int>
1 1-2         322
2 3           334
3 4           361
4 5           705
5 NA          720

# A tibble: 16 × 3
   `CKD stage` CKD_stage     n
         <dbl> <chr>     <int>
 1           0 1-2          97
 2           0 3            61
 3           0 4            12
 4           0 NA           22
 5           3 4           138
 6           3 5             9
 7           3 NA           39
 8           4 5            73
 9           4 NA           28
10           5 4             1
11           5 NA           95
12          12 1-2         134
13          12 3            69
14          12 4            15
15          12 5             1
16          12 NA           19

# A tibble: 2 × 2
  gender     n
   <dbl> <int>
1      1  1816
2      2   626

# A tibble: 8 × 2
  `year entry`     n
         <dbl> <int>
1         2007   160
2         2008   498
3         2009   400
4         2010   567
5         2011   470
6         2012   283
7         2014    59
8         2015     5

# A tibble: 12 × 2
   basic_diagnosis                  n
   <chr>                        <int>
 1 Chronic Glomerulo-nephritis    222
 2 Congenital disease               7
 3 Cystic disease                  22
 4 Diabetic Nephropathy           200
 5 Heredofamilial                   4
 6 Hypertensive Nephrosclerosis   123
 7 Obstructive uropathy            42
 8 Other                          191
 9 Renovascular disease             5
10 Tubulo interstitial disease     71
11 Undetermined                   122
12 <NA>                          1433

# A tibble: 8 × 2
  basic_diagnosis_pediatric                     n
  <chr>                                     <int>
1 Chronic Glomerulo-nephritis                 417
2 Cystic disease                               14
3 Dysplastic kidneys                           26
4 Metabolic diseases                           11
5 Obstructive uropathy                         84
6 Other                                       711
7 Tubulo interstitial disease including VUR   148
8 <NA>                                       1031

# A tibble: 15 × 2
   diagnosis_clean                               n
   <chr>                                     <int>
 1 Other                                       901
 2 Chronic Glomerulo-nephritis                 636
 3 Diabetic Nephropathy                        200
 4 Tubulo interstitial disease including VUR   143
 5 Obstructive uropathy                        125
 6 Hypertensive Nephrosclerosis                123
 7 Undetermined                                122
 8 Tubulo interstitial disease                  71
 9 Cystic disease                               36
10 <NA>                                         32
11 Dysplastic kidneys                           26
12 Metabolic diseases                           11
13 Congenital disease                            7
14 Renovascular disease                          5
15 Heredofamilial                                4

# A tibble: 15 × 2
   Etiologynew                                   n
   <chr>                                     <int>
 1 Other                                       901
 2 Chronic Glomerulo-nephritis                 636
 3 Diabetic Nephropathy                        200
 4 Tubulo interstitial disease including VUR   143
 5 Obstructive uropathy                        125
 6 Hypertensive Nephrosclerosis                123
 7 Undetermined                                122
 8 Tubulo interstitial disease                  71
 9 Cystic disease                               36
10 <NA>                                         32
11 Dysplastic kidneys                           26
12 Metabolic diseases                           11
13 Congenital disease                            7
14 Renovascular disease                          5
15 Heredofamilial                                4

# A tibble: 15 × 2
   Etiologynew                    n
   <chr>                      <int>
 1 Obstructive uropathy         515
 2 CGN                          333
 3 Undetermined                 329
 4 Hypoplasia-dysplasia         306
 5 Reflux nephropathy           290
 6 Neurogenic bladder           131
 7 Cystic kidney disease        124
 8 Tubulointerstitial disease   112
 9 Hemolytic uremic syndrome     41
10 Chronic pyelonephritis        36
11 Inherited tubular disease     32
12 Other CAKUT                   27
13 Nephrolithiasis               18
14 Renovascular disorders        12
15 AKIsequelae                    9


2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 <NA> 
 160  499  400  567  470  283    2   61    9    8  243  433  318  238 1066


 1-2    3    4    5 <NA> 
 322  842 1092 1768  733

Characteristic	Main Registry, N = 2,442¹	Paediatric Registry, N = 2,315¹
age_new	10.0 (0.0, 16.0)	8.0 (4.0, 11.0)
Unknown	0	11
Gender
Female	626 (26%)	555 (24%)
Male	1,816 (74%)	1,760 (76%)
state_new
Andhra Pradesh	116 (4.8%)	58 (2.5%)
Arunachal Pradesh	1 (<0.1%)	0 (0%)
Assam	23 (0.9%)	19 (0.8%)
Bihar	113 (4.6%)	67 (2.9%)
Chhattisgarh	11 (0.5%)	17 (0.7%)
Goa	1 (<0.1%)	0 (0%)
Gujarat	223 (9.1%)	2 (<0.1%)
Haryana	9 (0.4%)	105 (4.5%)
Himachal Pradesh	1 (<0.1%)	1 (<0.1%)
Jammu and Kashmir	0 (0%)	8 (0.3%)
Jharkhand	40 (1.6%)	19 (0.8%)
Karnataka	17 (0.7%)	232 (10%)
Kerala	30 (1.2%)	51 (2.2%)
Madhya Pradesh	84 (3.4%)	92 (4.0%)
Maharashtra	239 (9.8%)	302 (13%)
Manipur	8 (0.3%)	1 (<0.1%)
Meghalaya	2 (<0.1%)	0 (0%)
Mizoram	1 (<0.1%)	0 (0%)
New Delhi	52 (2.1%)	251 (11%)
Odisha	59 (2.4%)	7 (0.3%)
Other Union Territories	4 (0.2%)	12 (0.5%)
Punjab	85 (3.5%)	8 (0.3%)
Rajasthan	49 (2.0%)	62 (2.7%)
Sikkim	1 (<0.1%)	0 (0%)
Tamil Nadu	241 (9.9%)	259 (11%)
Telangana	54 (2.2%)	3 (0.1%)
Tripura	10 (0.4%)	1 (<0.1%)
Unknown	0 (0%)	360 (16%)
Uttar Pradesh	692 (28%)	188 (8.1%)
Uttarakhand	5 (0.2%)	4 (0.2%)
West Bengal	271 (11%)	186 (8.0%)
year_new
2007	160 (6.6%)	0 (0%)
2008	498 (20%)	1 (<0.1%)
2009	400 (16%)	0 (0%)
2010	567 (23%)	0 (0%)
2011	470 (19%)	0 (0%)
2012	283 (12%)	0 (0%)
2013	0 (0%)	2 (0.2%)
2014	59 (2.4%)	2 (0.2%)
2015	5 (0.2%)	4 (0.3%)
2016	0 (0%)	8 (0.6%)
2017	0 (0%)	243 (19%)
2018	0 (0%)	433 (35%)
2019	0 (0%)	318 (25%)
2020	0 (0%)	238 (19%)
Unknown	0	1,066
CKDstagenew
1-2	322 (19%)	0 (0%)
3	334 (19%)	508 (22%)
4	361 (21%)	731 (32%)
5	705 (41%)	1,063 (46%)
Unknown	720	13
Etiologynew
AKIsequelae	0 (0%)	9 (0.4%)
Chronic Glomerulo-nephritis	636 (26%)	333 (14%)
Chronic pyelonephritis	0 (0%)	36 (1.6%)
Congenital disease	7 (0.3%)	0 (0%)
Cystic disease	36 (1.5%)	124 (5.4%)
Diabetic Nephropathy	200 (8.3%)	0 (0%)
Dysplastic kidneys	26 (1.1%)	0 (0%)
Hemolytic uremic syndrome	0 (0%)	41 (1.8%)
Heredofamilial	4 (0.2%)	0 (0%)
Hypertensive Nephrosclerosis	123 (5.1%)	0 (0%)
Hypoplasia-dysplasia	0 (0%)	306 (13%)
Inherited tubular disease	0 (0%)	32 (1.4%)
Metabolic diseases	11 (0.5%)	0 (0%)
Nephrolithiasis	0 (0%)	18 (0.8%)
Neurogenic bladder	0 (0%)	131 (5.7%)
Obstructive uropathy	125 (5.2%)	515 (22%)
Other	901 (37%)	0 (0%)
Other CAKUT	0 (0%)	27 (1.2%)
Reflux nephropathy	0 (0%)	290 (13%)
Renovascular disease	5 (0.2%)	12 (0.5%)
Tubulo interstitial disease	214 (8.9%)	112 (4.8%)
Undetermined	122 (5.1%)	329 (14%)
Unknown	32	0
Etiology_new
CAKUT	194 (8.0%)	1,262 (55%)
Chronic glomerulonephritis	636 (26%)	333 (14%)
Diabetes	200 (8.3%)	0 (0%)
Hypertension	123 (5.1%)	0 (0%)
Tubulo-interstitial disease	214 (8.9%)	144 (6.2%)
Other and unspecified	1,043 (43%)	576 (25%)
Unknown	32	0
¹ Median (IQR); n (%)

# check the missing data in etiology_new in the main registry


combined_data %>%
  filter(dataset == "Main Registry", is.na(Etiology_new)) %>%
  count(Etiologynew)

# A tibble: 1 × 2
  Etiologynew     n
  <chr>       <int>
1 <NA>           32

Recoding of the etiologies into broader and fewer categories was done to allow for easier comparison of the two datasets. The etiologies were recoded into the following categories: CAKUT, Chronic glomerulonephritis, Diabetes, Hypertension, Tubulo-interstitial disease, and Other and unspecified. The recoding was done as follows:

Characteristic	CAKUT, N = 1,456¹	Chronic glomerulonephritis, N = 969¹	Diabetes, N = 200¹	Hypertension, N = 123¹	Tubulo-interstitial disease, N = 358¹	Other and unspecified, N = 1,619¹
Etiologynew
AKIsequelae	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	9 (0.6%)
Chronic Glomerulo-nephritis	0 (0%)	969 (100%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Chronic pyelonephritis	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	36 (2.2%)
Congenital disease	7 (0.5%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Cystic disease	160 (11%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Diabetic Nephropathy	0 (0%)	0 (0%)	200 (100%)	0 (0%)	0 (0%)	0 (0%)
Dysplastic kidneys	26 (1.8%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Hemolytic uremic syndrome	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	41 (2.5%)
Heredofamilial	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	4 (0.2%)
Hypertensive Nephrosclerosis	0 (0%)	0 (0%)	0 (0%)	123 (100%)	0 (0%)	0 (0%)
Hypoplasia-dysplasia	306 (21%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Inherited tubular disease	0 (0%)	0 (0%)	0 (0%)	0 (0%)	32 (8.9%)	0 (0%)
Metabolic diseases	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	11 (0.7%)
Nephrolithiasis	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	18 (1.1%)
Neurogenic bladder	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	131 (8.1%)
Obstructive uropathy	640 (44%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Other	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	901 (56%)
Other CAKUT	27 (1.9%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Reflux nephropathy	290 (20%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Renovascular disease	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	17 (1.1%)
Tubulo interstitial disease	0 (0%)	0 (0%)	0 (0%)	0 (0%)	326 (91%)	0 (0%)
Undetermined	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	451 (28%)
¹ n (%)

Ages 0 and >15 are over-represented in the main registry.

Age group 15-18 are over-represented in main registry.

Sex distribution is the same in both registries.

UP, GJ, WB and AP are over-represented in the main registry.The top states in main registry are UP, WB, TN and MH. The top states in the paediatic registry are MH, TN, ND and KA, with a large proportion of unknown state.

The years of admission have very minimal overlap between the two registries. However, the paediatric registry has a high proportion of missing years (>20%).

Paediatric registry does not have cases in Stage 1-2. Could this be real? Are we missing something. Could the NA cases be in stage 1-2? In both datasets the stage 5 is the most common.

The most common diagnosis in main registry is Other and unspecified whereas in the paediatric registry it is CAKUT. Diabetes and hypertension as a cause are available only in the main registry.

Across most CKD stages the common etiology was Other and unspecified in the main, but CAKUT in the paediatric registry.

The proportion of CAKUT causes are declining with increasing age in the paediatric registry. The proportion of Other and unspecified causes are increasing with age in the main registry. DM and HTN are present only in 0-5 age group in the main registry.

There does not seem to be any difference in etiology distribution of CKD stages across age groups in the main registry. In the paediatric registry, the proportion of CAKUT is higher in the 0-5 age group compared to the other age groups.

The proportion of CAKUT across ages is decreasing as per the paediatric registry.

In both registries, the proportion of other and unspecified causes have increased over the years. The proportion of CAKUT has decreased in the paediatric registry. DM and HTN appear to increase over the years in the main registry.

Table of Prevalence fractions across age groups

age_new	Diabetes	Hypertension	Chronic glomerulonephritis	Other and unspecified
<1 year	6.78	0.00	27.73	65.50
1 year	7.47	0.00	25.73	66.80
2-4 years	7.84	0.00	23.96	68.19
5-9 years	9.07	0.00	17.11	73.82
10-14 years	8.94	0.00	7.61	83.45
15-19 years	10.75	0.83	3.72	84.70
20-24 years	12.02	0.78	2.76	84.44
25-29 years	13.63	0.92	2.50	82.95

Table of YLD fractions across age groups

age_new	Diabetes	Hypertension	Chronic glomerulonephritis	Other and unspecified
<1 year	1.67	0.00	18.42	79.91
1 year	1.73	0.00	18.26	80.01
2-4 years	1.84	0.00	18.63	79.53
5-9 years	2.98	0.00	20.83	76.19
10-14 years	2.68	0.00	19.95	77.37
15-19 years	1.55	13.38	18.95	66.12
20-24 years	2.20	14.91	21.42	61.47
25-29 years	3.66	16.84	23.00	56.50

Comparison of GBD prevalence fractions with registry data - overall

In this graph, we are comparing the etiology fractions of GBD prevalence number for the years 2020 grouped within age group with the two registry data sources.

We can see here that the registry data is not consistent with the GBD data. The registry data has a higher fraction of Hypertension and Diabetes compared to the GBD data in the <1 year age. Both the registry data has an increasing fraction of Chronic glomerulonephritis with age as opposed to the GBD data. The fraction of other and unspecified etiology is the opposite direction. The fraction of DM and HTN are beginning to make an impression only after age 15 in the registry data unlike GBD where it starts from the beginning (most of it is type 1 DM).

Comparison of GBD YLD fractions with registry data - overall

In this graph, we are comparing the etiology fractions of GBD YLD number for the years 2020 grouped within age group with the two registry data sources.

Interestingly, the YLD etiology fractions are similar to the registry etiology fractions unlike the prevalence fractions.

Comparison of GBD prevalence fractions with registry data - statewise

In the above graphs, the states with most cases in the registry have been plotted. There does not seem to be a consistent pattern within any one particular state. However, the states figures generally mirror the national figure.

Comparison of GBD YLD fractions with registry data - statewise

There does not seem to be a consistent pattern within any one particular state. However, the states figures generally mirror the national figure.