Hugging Face's logo

Datasets:
electricsheepafrica
/
africa-synth-sickle-cell-disease-all

Tasks:
Tabular Classification
Tabular Regression
Modalities:
Tabular
Text
Formats:
csv
Languages:
English
Size:
10K - 100K
Tags:
Synthetic
healthcare
sickle-cell-disease
hematology
hemoglobinopathy
hydroxyurea
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
Dataset Viewer
Auto-converted to Parquet Duplicate
id
int64
age_years
float64
sex
string
genotype
string
on_hydroxyurea
int64
hu_dose_mg_per_kg
float64
hemoglobin_gdl
float64
hbf_pct
float64
wbc_10e3_per_ul
float64
platelets_10e3_per_ul
int64
reticulocyte_pct
float64
ldh_ul
int64
total_bilirubin_mgdl
float64
voc_episodes_per_year
int64
transfusions_per_year
int64
pain_score_nrs
int64
spleen_status
string
hx_acute_chest_syndrome
int64
hx_stroke
int64
hx_splenic_sequestration
int64
hx_avascular_necrosis
int64
hx_priapism
int64
hx_leg_ulcers
int64
hx_gallstones
int64
hx_nephropathy
int64
hx_retinopathy
int64
hx_pulmonary_hypertension
int64
outcome
string
1
14.6
F
HbSS
1
12.5
9.1
13
15.5
238
11.3
100
2.3
0
1
1
auto_splenectomy
0
0
0
0
0
0
1
0
1
0
alive
2
29.9
F
HbSS
1
17.1
6.5
15.1
3.4
341
10.2
370
2.5
2
0
3
auto_splenectomy
0
0
0
1
0
0
0
1
0
0
alive
3
2
M
HbS_beta_thal
0
0
8.3
5.8
15.2
245
8.6
439
0.3
2
2
5
intact
0
0
0
0
0
0
0
0
0
0
alive
4
16.7
M
HbSS
1
16.2
8.5
16.1
8.2
568
13.4
347
4
3
0
7
auto_splenectomy
0
1
0
0
1
0
1
1
0
0
alive
5
1.2
F
HbSS
1
14
8.8
17.2
5.4
214
10.4
601
2.6
0
1
1
intact
0
0
0
0
0
0
0
0
0
0
alive
6
4.8
M
HbSS
1
16.5
9.6
24
17
531
8.2
233
2.5
2
0
3
intact
0
0
0
0
0
0
0
0
0
0
alive
7
16.9
F
HbSS
1
19
8.2
13.4
8.7
187
11.2
826
6.6
1
0
4
auto_splenectomy
0
0
1
0
0
1
0
1
0
0
alive
8
6.6
F
HbSC
1
14.7
11.7
11.2
6.8
316
5
276
2.5
1
0
4
splenectomy
0
0
1
0
0
0
0
0
0
0
alive
9
5.4
F
HbSS
1
20.2
8.6
12.8
9.7
559
11.2
399
4.4
1
1
5
intact
0
0
0
0
0
0
0
0
0
0
alive
10
28.7
M
HbSC
1
15.9
12.4
10.9
4.6
119
0.5
235
1.4
3
0
8
splenectomy
0
0
1
0
1
1
1
0
1
0
alive
11
14.1
M
HbSC
0
0
12.4
4.6
9.6
100
4.6
135
2.2
1
0
5
intact
0
0
0
0
0
0
1
0
1
0
alive
12
52.9
F
HbSS
1
17.7
9.5
16.8
14.1
469
8.7
471
1
0
1
0
auto_splenectomy
1
0
0
1
0
0
1
0
1
0
alive
13
13.2
M
HbSC
0
0
10
1
6
191
0.5
398
1.7
0
0
1
intact
0
1
0
0
0
0
0
0
0
0
alive
14
5.3
F
HbSS
1
16.3
9.9
20.5
12.6
441
8.9
761
2.2
2
0
3
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
15
4
M
HbSS
1
16.7
7.8
15.1
19.5
255
13.5
227
3.2
1
0
3
intact
0
0
0
0
0
0
1
0
0
0
alive
16
17.8
M
HbSS
1
17.2
8.1
11.4
7.4
403
7
307
4.3
1
0
5
auto_splenectomy
0
0
0
1
1
0
1
0
0
0
alive
17
3.1
M
HbSS
0
0
7
9.7
17.4
581
1
479
0.5
5
2
6
intact
0
0
0
0
0
0
1
0
0
0
alive
18
6.3
M
HbSS
1
18.8
11
17.3
14.8
256
6.9
720
3.8
1
1
4
intact
0
0
0
0
0
0
0
0
0
0
alive
19
1.2
F
HbSS
1
17.3
10
12.3
9.4
404
11
584
1
2
0
4
intact
0
0
0
0
0
0
0
0
0
0
alive
20
4
F
HbSS
1
18.9
6.6
18.2
3.4
222
8.4
345
3.3
0
2
2
intact
0
0
1
0
0
0
0
0
0
0
alive
21
10.8
F
HbSC
0
0
10.4
1.4
6.7
352
4.8
325
1.5
1
0
2
splenectomy
0
0
1
0
0
0
1
0
1
0
alive
22
13.9
F
HbSS
1
18
7.2
13.7
3
255
5.8
610
2.1
0
0
2
auto_splenectomy
0
0
0
0
0
0
1
0
0
0
alive
23
31.7
M
HbSS
0
0
8.5
6.7
13.2
376
12.8
741
1.5
1
1
2
auto_splenectomy
1
1
0
1
0
1
1
0
0
0
alive
24
2.9
M
HbS_beta_thal
1
18.3
11.5
11.1
7.7
219
9.8
241
3.1
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
25
12.8
F
HbSS
1
17.1
5.5
20.6
9.5
289
5.5
487
2.5
0
0
2
auto_splenectomy
0
0
0
0
0
0
1
0
0
0
alive
26
2
M
HbS_beta_thal
0
0
5.8
2.9
13.3
276
4.6
407
2.8
0
2
0
intact
0
0
0
0
0
0
0
0
0
0
alive
27
16
M
HbSC
1
19.7
12.3
12
10.5
211
1.8
387
0.9
0
0
2
intact
0
0
1
0
0
0
0
0
0
0
alive
28
16.7
M
HbSS
1
17.9
7.8
8.5
18.5
609
7.6
653
4
2
0
3
splenectomy
1
0
1
0
0
0
0
0
0
0
alive
29
9
F
HbSS
1
16.4
8.6
22.6
4.7
313
1
298
0.6
1
1
3
auto_splenectomy
1
0
0
0
0
0
0
0
0
0
alive
30
0.7
M
HbSC
0
0
7.2
2
10.1
197
2.9
227
1.6
2
2
4
intact
0
0
0
0
0
0
0
0
0
0
alive
31
14.8
M
HbSC
0
0
9.6
1
11.8
180
1.5
374
0.2
1
0
4
splenectomy
1
0
1
0
0
0
0
0
0
0
alive
32
10.4
F
HbSC
0
0
12.3
0.9
9.2
170
4
248
0.7
1
0
4
intact
1
0
0
0
0
0
0
0
0
0
alive
33
15.3
F
HbSC
0
0
8.1
1.9
10.4
304
2.6
228
1.4
2
0
5
intact
1
0
1
0
0
0
0
0
0
0
alive
34
7.1
M
HbSS
1
20.4
8.4
16.5
8.4
290
7.2
100
2.4
0
0
0
intact
0
0
0
0
0
0
1
0
0
0
alive
35
1.2
F
HbSS
1
16.6
8.9
23.5
15.7
349
3.5
390
1.9
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
36
3.2
M
HbS_beta_thal
0
0
5.9
2.2
7.9
167
9.8
491
2.1
3
3
4
intact
0
0
0
0
0
0
0
0
0
0
alive
37
22
M
HbSS
1
16.9
8.8
18
7.1
571
5.2
166
3.2
0
0
1
auto_splenectomy
0
0
0
1
0
0
1
0
1
0
alive
38
0.9
M
HbSS
0
0
7.7
6.5
18.1
403
6.7
381
3.2
3
1
5
intact
0
0
0
0
0
0
0
0
0
0
alive
39
4
M
HbSS
0
0
7.1
14.1
16.3
176
9.3
251
5.2
1
0
5
intact
0
0
0
0
0
0
1
0
0
0
alive
40
10.5
M
HbSC
0
0
9
3.4
6.9
92
1.5
275
2
0
0
1
intact
0
0
0
0
0
0
0
0
0
0
alive
41
14.4
F
HbSC
0
0
11.3
0.3
9.9
326
5
264
1.5
1
0
5
intact
1
0
0
0
0
0
0
0
0
0
alive
42
20.8
M
HbSS
0
0
7.7
15
20.2
464
10.2
494
3.1
5
0
5
auto_splenectomy
1
0
0
0
0
1
0
0
0
0
alive
43
0.6
F
HbSS
0
0
7.3
5.4
21.4
277
15.2
100
2.6
3
2
7
intact
0
0
0
0
0
0
0
0
0
0
alive
44
14.8
M
HbSS
1
16.4
7.9
17.6
8.7
518
7.3
294
1.6
5
0
6
auto_splenectomy
1
1
0
1
0
0
1
0
0
0
alive
45
3.5
M
HbSS
1
14.7
7.2
20.1
14.4
181
5.1
287
2.6
0
0
1
intact
0
0
0
0
0
0
0
0
0
0
alive
46
4.1
F
HbS_beta_thal
1
15.9
10.3
15.2
10.3
360
3.5
208
1.5
1
0
3
intact
0
0
0
0
0
0
0
0
0
0
alive
47
7.7
F
HbSS
1
18.4
9.4
19.5
10.7
461
8.9
434
0.6
0
0
0
intact
0
0
0
0
0
0
1
0
0
0
alive
48
3.9
M
HbS_beta_thal
1
18.4
10.8
14.1
6.8
266
9.8
549
1.8
1
1
4
intact
0
0
0
0
0
0
0
0
0
0
alive
49
2.3
M
HbSS
1
17
8.5
22.6
11.9
353
1
749
1.5
0
0
1
intact
0
0
0
0
0
0
0
0
0
0
alive
50
1.9
F
HbSC
0
0
7.2
0.2
6.4
289
6.3
334
2.9
0
0
1
intact
0
0
0
0
0
0
0
0
0
0
alive
51
29.9
M
HbSS
1
18.7
7.8
15.3
3
263
1
232
3.9
0
0
0
auto_splenectomy
0
0
0
1
0
1
1
1
0
0
alive
52
4.9
F
HbSC
0
0
7.5
0.2
3.7
312
7.3
293
2.3
2
0
3
intact
0
0
0
0
0
0
0
0
0
0
alive
53
16.3
F
HbSS
1
19
6.8
22.6
7.2
424
8.2
116
1.7
1
0
4
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
54
6.9
M
HbSC
0
0
10.8
0.2
9.9
137
5.1
164
1.9
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
55
1.9
F
HbSC
0
0
11.2
2.7
8.8
369
2.4
247
2
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
56
6.4
M
HbSS
1
15.6
8.9
15.3
4.9
464
3.3
956
1.2
1
0
3
intact
0
1
1
0
0
0
0
0
0
0
alive
57
3.7
F
HbS_beta_thal
0
0
6.3
7.4
12.4
213
12.1
449
1.9
3
0
5
intact
0
0
0
0
0
0
0
0
0
0
alive
58
4.6
F
HbSS
1
16.7
8.4
22.6
17.6
363
3.2
248
0.9
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
59
16.3
F
HbSS
1
20.5
7.5
16.7
11.3
484
1
498
3.7
1
0
2
auto_splenectomy
1
0
0
1
0
0
1
0
0
0
alive
60
9.4
F
HbSS
0
0
3.8
10.3
24.8
481
11.2
445
3.3
0
2
0
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
61
9.5
F
HbSS
1
17.5
10.3
14.8
15.7
398
14.8
718
3.3
1
1
3
auto_splenectomy
0
0
0
0
0
0
0
1
0
0
alive
62
1.1
M
HbSC
0
0
10.2
2.4
11.9
203
4.2
281
0.4
2
0
4
intact
0
0
0
0
0
0
0
0
0
0
alive
63
1.2
M
HbSC
0
0
10.3
0.2
7.7
303
4.3
315
0.8
0
0
1
intact
0
0
0
0
0
0
0
0
0
0
alive
64
2.4
M
HbSS
1
17.8
9.2
12.1
13.6
506
2.2
419
2.3
1
0
3
intact
1
0
0
0
0
0
0
0
0
0
alive
65
1.4
M
HbSC
0
0
10.4
1.1
5.5
246
2.9
285
2.2
1
1
4
intact
0
0
0
0
0
0
0
0
0
0
alive
66
4.8
F
HbSS
0
0
4.9
7.4
9.8
383
7.7
264
3.6
4
2
5
auto_splenectomy
1
0
0
0
0
0
0
0
0
0
died
67
7.4
M
HbSS
1
15.9
11
16
7.1
243
12
325
4.9
1
0
4
splenectomy
1
0
1
0
0
0
0
0
0
0
died
68
5.2
F
HbSC
1
16.2
8.9
11.9
8.9
268
3.7
327
2.4
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
69
44.6
M
HbSS
1
17.2
7.1
13.3
8.6
530
10.2
636
2.7
1
0
5
auto_splenectomy
1
1
0
0
1
1
0
1
1
0
alive
70
12.5
M
HbSS
0
0
8.6
0.5
10.7
352
8.1
436
3.8
0
0
1
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
71
8.4
M
HbSS
1
21.5
8.7
22.5
3
531
5.7
763
2.4
0
2
1
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
72
46.4
F
HbSS
0
0
8
2.6
7
456
1
257
4.2
1
1
3
auto_splenectomy
1
1
0
1
0
1
1
1
1
0
alive
73
6.9
M
HbSS
1
15.7
7.8
12.1
17.2
581
8.3
755
5.3
2
0
3
intact
0
0
0
0
0
0
1
0
0
0
died
74
3.6
M
HbSS
0
0
7
5.4
12.9
191
12.3
434
3.1
5
1
7
intact
0
0
0
0
0
0
1
0
0
0
alive
75
5.2
M
HbSS
1
17.8
8.4
12.3
10.2
161
6.4
277
1.7
1
0
3
intact
0
0
0
0
0
0
0
0
0
0
alive
76
33.2
F
HbSC
1
17.9
8
10.6
8.3
222
4.5
205
1.1
0
0
2
intact
0
0
1
0
0
0
1
1
1
0
alive
77
1.3
M
HbSS
1
16.4
9
16.1
11.2
612
7.3
403
1.5
1
1
3
intact
0
0
0
0
0
0
0
0
0
0
alive
78
1.5
F
HbSC
0
0
9.6
5
10.7
365
4.7
100
2
0
0
2
intact
0
0
0
0
0
0
0
0
0
0
alive
79
4
M
HbSC
1
17.5
11
11.5
10.5
186
3.5
192
0.3
1
0
3
splenectomy
0
0
1
0
0
0
0
0
0
0
alive
80
16.6
M
HbSS
1
19.4
7.6
14.9
12.8
195
14.3
307
2.9
0
1
1
auto_splenectomy
1
0
0
1
1
0
0
1
1
0
alive
81
1.9
F
HbSS
0
0
7
9.5
14.5
674
7.6
563
2.9
6
1
9
intact
0
0
0
0
0
0
0
0
0
0
died
82
2.1
F
HbSS
1
19.6
9.1
18.7
11.4
143
6.5
305
2.3
1
1
3
intact
0
0
0
0
0
0
0
0
0
0
alive
83
7.6
F
HbSC
1
16.4
10.1
10.5
5.6
176
1.7
207
1.5
0
0
0
intact
0
0
0
0
0
0
0
0
0
0
alive
84
31.7
M
HbSC
0
0
9.1
1.8
8.8
80
9
352
1.8
1
0
2
splenectomy
1
0
1
0
1
1
1
0
1
0
alive
85
4.9
F
HbSS
0
0
6.3
8.5
7.1
444
3.3
246
1.9
8
1
9
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
86
3.5
F
HbSS
0
0
9
2.5
14
112
10.1
847
3.8
1
1
3
splenectomy
0
0
1
0
0
0
1
0
0
0
alive
87
4.8
F
HbSS
0
0
4.3
8.2
10.4
539
9.6
212
2.8
3
3
4
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
88
2.1
M
HbSS
1
15.6
6.9
5.7
9.6
338
9.8
704
2.6
1
0
4
intact
0
0
0
0
0
0
0
0
0
0
alive
89
11.9
M
HbSS
1
16.3
6.4
12.3
10.3
495
8.3
474
2.6
0
1
1
splenectomy
0
0
1
0
0
0
0
1
0
0
alive
90
30.7
M
HbSS
1
17.4
8.3
26.1
10
303
8.5
365
0.7
1
0
3
auto_splenectomy
1
0
0
0
1
0
1
0
1
0
alive
91
33.2
F
HbSS
0
0
5.2
4.4
17.8
532
5.5
468
3.2
2
4
5
splenectomy
1
0
1
1
0
0
1
0
0
0
alive
92
7.2
F
HbSC
0
0
9.4
5.6
11.3
279
4.1
144
2.6
1
0
2
intact
0
0
0
0
0
0
0
0
0
0
alive
93
16
M
HbSS
1
16.4
9.1
18.5
7.1
253
10.3
720
5.6
0
0
1
auto_splenectomy
0
0
0
0
0
0
0
1
0
0
alive
94
2.3
M
HbSS
1
18.6
6.9
15.7
3
80
10.4
329
0.3
2
1
2
intact
0
0
0
0
0
0
0
0
0
0
alive
95
13.6
M
HbSS
1
20.9
6.8
18.2
12.1
125
13.2
361
3.4
0
2
1
auto_splenectomy
0
0
0
0
0
0
0
1
0
0
alive
96
39.8
M
HbSC
0
0
10.6
3
6.2
248
4.5
280
1.6
0
0
0
intact
1
0
1
0
1
0
1
1
0
1
alive
97
1.7
F
HbSS
1
15.1
9.3
14.1
13.3
310
14
219
5
1
2
4
intact
0
0
0
0
0
0
0
0
0
0
died
98
2.5
M
HbSS
0
0
6.3
18.6
14.5
491
9.4
636
4
2
0
5
intact
0
0
0
0
0
0
0
0
0
0
alive
99
11.1
M
HbSS
1
16.7
8.2
20.3
3
254
8.4
600
4.5
3
2
7
auto_splenectomy
0
0
0
0
0
0
0
0
0
0
alive
100
2.9
M
HbSC
0
0
9.6
2.2
11.6
342
5.5
285
3.5
3
0
6
intact
0
0
0
0
0
0
0
0
0
0
alive
End of preview. Expand in Data Studio

⚠️ Synthetic dataset — Parameterized from published SSA literature, not real observations. Not suitable for empirical analysis or policy inference.

Sickle Cell Disease Registry Dataset

Abstract

This dataset provides 30,000 synthetic registry records (10,000 per scenario) of patients with sickle cell disease (SCD) in sub-Saharan African clinical settings. Each record contains 28 variables spanning genotype (HbSS, HbSC, HbS-β-thalassemia), steady-state laboratory values (hemoglobin, HbF, WBC, platelets, reticulocytes, LDH, bilirubin), hydroxyurea therapy status and dosing, annual vaso-occlusive crisis frequency, transfusion history, 10 chronic complications (stroke, acute chest syndrome, splenic sequestration, avascular necrosis, priapism, leg ulcers, gallstones, nephropathy, retinopathy, pulmonary hypertension), spleen status, pain scores, and survival outcome. Three care-level scenarios span the spectrum from comprehensive urban tertiary care (3.3% mortality) to limited rural settings (16.3% mortality), reflecting the massive survival disparities across sub-Saharan Africa.

1. Introduction

Sickle cell disease is the most common life-threatening genetic disorder globally, with an estimated 300,000–400,000 affected neonates born annually, over 75% in sub-Saharan Africa (Ware et al., Lancet 2017). Despite this enormous burden, structured open-access registry data for SCD in Africa is virtually non-existent. In settings without newborn screening or hydroxyurea access, 50–90% of children with SCA die before age 5 (StatPearls 2024). The REACH trial (Tshimanga et al., NEJM 2018) demonstrated that hydroxyurea is feasible and reduces VOC by 55%, infections by 38%, and mortality by 70% in sub-Saharan children — yet access remains extremely limited.

This dataset is entirely synthetic. It must not be used for clinical decision-making.

2. Methodology

2.1 Target Population

SCD patients aged 0.5–55 years enrolled in facility-based registries across sub-Saharan Africa. Age distribution is skewed toward children and young adults, reflecting the survival bias inherent in SSA SCD populations.

2.2 Epidemiological Parameterization

All parameters are sourced from peer-reviewed literature:

Parameter Value Source
SCA (HbSS) prevalence in Africa 1.43% (95% CI 1.08–1.88%) PLOS One 2024 meta-analysis (n=1.2M)
HbSS proportion among SCD ≥70% globally Ware et al., Lancet 2017
HbSC proportion ~25–30% StatPearls 2024; regional variation
Mortality (hospital cohort, Tanzania) 1.9/100 PYO overall; 7.3/100 PYO <5y Makani et al. 2011
HU effect on VOC IRR 0.45 (98.3→44.6/100 PYO) REACH Trial, NEJM 2018
HU effect on mortality IRR 0.30 (3.6→1.1/100 PYO) REACH Trial, NEJM 2018
HU dose 17.5 ± 1.8 mg/kg/day REACH Trial, NEJM 2018
HU retention at 3 years 94.2% REACH Trial, NEJM 2018
Stroke incidence (HbSS) 0.61/100 PYO; 11% by age 20 Ohene-Frempong et al., Blood 1998
ACS incidence (children) 12.8/100 PYO Castro et al., Blood 1994
Priapism (HbSS males, lifetime) ~35% Platt et al., NEJM 1991
CSSCD median survival (HbSS) Men 42y, Women 48y (US) Platt et al., NEJM 1994
Under-5 mortality (SSA, no treatment) 50–90% StatPearls 2024; multiple sources
Risk factor: Hb <5 g/dL OR 3.8 for death Makani et al. 2011

2.3 Laboratory Reference Values

Parameter HbSS (mean ± SD) HbSC (mean ± SD) HbS-β-thal (mean ± SD) Source
Hemoglobin (g/dL) 7.5 ± 1.3 10.5 ± 1.5 8.5 ± 1.5 Antwi-Boasiako 2018; Kato 2022
HbF (%) 8.0 ± 5.0 3.0 ± 2.0 6.0 ± 4.0 Antwi-Boasiako 2018
WBC (×10³/µL) 12.5 ± 4.5 9.0 ± 3.0 10.5 ± 3.5 Antwi-Boasiako 2018
Platelets (×10³/µL) 380 ± 130 250 ± 80 320 ± 110 Antwi-Boasiako 2018
Reticulocyte (%) 10.0 ± 4.0 4.0 ± 2.0 7.0 ± 3.0 Literature consensus
LDH (U/L) 500 ± 200 300 ± 100 400 ± 150 Literature consensus
Total Bilirubin (mg/dL) 3.0 ± 1.5 1.5 ± 0.8 2.2 ± 1.2 Literature consensus

2.4 Scenario Design

Scenario Setting HU Uptake Mortality Mean VOC/yr Deaths
Comprehensive care Urban tertiary (Korle-Bu, MNH) 70% 3.3% 1.4 334
Moderate care District hospital (regional Nigeria/DRC) 30% 7.8% 1.9 776
Limited care Rural, no screening, minimal HU 5% 16.3% 2.3 1,627

2.5 Generation Process

Sampling follows a directed acyclic graph: demographics → genotype → HU status → lab values (conditional on genotype + HU) → VOC frequency (conditional on Hb, HbF, genotype, HU) → transfusions → complications (cumulative probability based on age-adjusted annualized rates) → spleen status → outcome (conditional on severity markers).

3. Schema

Column Type Description
id int Unique identifier
age_years float Age (0.5–55)
sex categorical M / F
genotype categorical HbSS / HbSC / HbS_beta_thal
on_hydroxyurea binary Currently on HU therapy
hu_dose_mg_per_kg float HU dose (mg/kg/day; 0 if not on HU)
hemoglobin_gdl float Steady-state hemoglobin (g/dL)
hbf_pct float Fetal hemoglobin (%)
wbc_10e3_per_ul float White blood cell count (×10³/µL)
platelets_10e3_per_ul int Platelet count (×10³/µL)
reticulocyte_pct float Reticulocyte percentage
ldh_ul int Lactate dehydrogenase (U/L)
total_bilirubin_mgdl float Total bilirubin (mg/dL)
voc_episodes_per_year int Annual vaso-occlusive crises
transfusions_per_year int Annual blood transfusions
pain_score_nrs int Pain numeric rating scale (0–10)
spleen_status categorical intact / auto_splenectomy / hyposplenism / splenectomy
hx_acute_chest_syndrome binary Ever diagnosed with ACS
hx_stroke binary Ever had stroke
hx_splenic_sequestration binary Ever had splenic sequestration
hx_avascular_necrosis binary Ever had AVN
hx_priapism binary Ever had priapism (males only)
hx_leg_ulcers binary Ever had leg ulcers
hx_gallstones binary Ever had gallstones
hx_nephropathy binary Ever diagnosed with nephropathy
hx_retinopathy binary Ever diagnosed with retinopathy
hx_pulmonary_hypertension binary Ever diagnosed with pulmonary HTN
outcome categorical alive / died

4. Validation

Validation Report

Key validation checks:

  • Cross-scenario monotonicity: Mortality increases from 3.3% → 7.8% → 16.3% as care level decreases ✓
  • Genotype-lab concordance: HbSS Hb ~7.5, HbSC ~10.5 g/dL matching literature ✓
  • HU effect: HbF significantly higher in HU-treated HbSS patients ✓
  • Complication accumulation: Prevalence increases with age as expected ✓
  • VOC frequency: Higher in HbSS, reduced by HU (consistent with REACH IRR 0.45) ✓

5. Usage 

from datasets import load_dataset
dataset = load_dataset("electricsheepafrica/sickle-cell-disease-registry", "moderate_care")
df = dataset["train"].to_pandas()

import pandas as pd
df = pd.read_csv("data/scd_moderate_care.csv")

Regenerate with custom parameters:

python generate_dataset.py --scenario limited_care --n 20000 --seed 123
python generate_dataset.py --all-scenarios --n 10000 --seed 42

6. Limitations

  • Synthetic: Not derived from real patient records. Must not be used for clinical decisions, treatment planning, or diagnostic validation.
  • Simplified genotype model: Only three genotypes modelled; does not include rare compound heterozygotes (HbSD, HbSO-Arab, etc.).
  • No longitudinal trajectories: Each record is a cross-sectional snapshot; no time-series of crises, lab values, or disease progression.
  • No co-morbidities: Does not model HIV, malaria, or malnutrition co-occurrence, which significantly impact SCD outcomes in SSA.
  • No geographic/ethnic variation: HbF levels vary by β-globin haplotype (Senegal, Benin, CAR, Arab-Indian); this is not modelled.
  • Complication independence: Complications are sampled independently; real-world clustering (e.g., stroke + ACS) is not captured.
  • HU adherence simplified: Binary on/off with a single adherence factor; real-world adherence is continuous and time-varying.

7. References

  1. PLOS One (2024). Burden of sickle cell anemia in Africa: systematic review & meta-analysis. 115 studies, n=1,203,839. DOI: 10.1371/journal.pone.0337090
  2. Makani J, et al. (2011). Mortality in SCA in Africa: prospective cohort in Tanzania. PLoS ONE, 6(2):e14699. PMID: 21358818
  3. Tshimanga MI, et al. (2018). Hydroxyurea for children with SCA in sub-Saharan Africa (REACH). NEJM, 379(18):1699-1710. PMID: 30501550
  4. Antwi-Boasiako C, et al. (2018). Hematological parameters in Ghanaian SCD patients. J Blood Med, 9:203-209. DOI: 10.2147/JBM.S170361
  5. Kato GJ, et al. (2022). Sickle Cell Disease: A Review. JAMA, 328(1):57-68. PMID: 35788790
  6. Ware RE, et al. (2017). Sickle cell disease. Lancet, 390(10091):311-323. PMID: 29542687
  7. Platt OS, et al. (1994). Mortality in SCD — CSSCD. NEJM, 330(23):1639-1644. PMID: 7993409
  8. Platt OS, et al. (1991). Pain in SCD: rates and risk factors. NEJM, 325(1):11-16. PMID: 2023281
  9. Ohene-Frempong K, et al. (1998). Cerebrovascular accidents in SCD. Blood, 91(1):288-294. PMID: 9414296
  10. Castro O, et al. (1994). ACS in SCD: incidence & risk factors. Blood, 84(2):643-649. PMID: 8043556
  11. de Montalembert M, et al. (2020). HbSS vs HbSC: distinction of genotypes. PLoS ONE, 15(1):e0228399. PMID: 31995624
  12. Zacharia A, et al. (2021). SLR of frequency of VOCs in SCD. Ann Hematol, 101:1-15. PMID: 34727959
  13. StatPearls (2024). Sickle Cell Anemia. NCBI Bookshelf NBK482164.

Citation 

@dataset{esa_scd_registry_2025,
  title={Sickle Cell Disease Registry Dataset},
  author={Electric Sheep Africa},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/electricsheepafrica/sickle-cell-disease-registry}
}

License

CC-BY-4.0
Downloads last month
76
Number of rows:
30,000
Total file size:
3.26 MB