Western Kenya Cancer Statistics

Western Kenya Cancer Statistics

Authors: ECR Group

1 Background

Eldoret Cancer Registry was established in 1999. It is located in the Faculty of Medical Sciences of Moi University, Department of Haematology and Blood Transfusion. The registry records details of all cancer patients diagnosed and treated in hospitals of Eldoret town, and aims to be population-based for the district of Uasin Gishu, in the Rift Valley Region of Kenya (pop. 2009: 894,179). As well as the Director (Dr N. Buziba), the registry has three registrars, Gladys Chesumbai , Jacqueline Gavana, and Jane Chepkosgei. Because of shortage of funds in the last 6 years, the staff work only part time basis on registry duties. Case-finding uses three types of source: hospital records departments, pathology and death

2 Datasource

The largest hospital is Moi Teaching and Referral Hospital which also acts as the Province General Hospital. Cases are identified from the disease index (completed by records staff in ICD-10) and details of cases abstracted from case records onto a registry form. The registrars also visits private hospitals (Eldoret, Elgon View, Medi-Heal Fertility Hospital, St. Lukes and Reale Hospital) and the Eldoret Hospice at regular intervals to identify cancer cases, although because of financial constraints, visits were widely spaced in recent years. There is only one public pathology lab. Serving the District (in Moi Teaching and Referral Hospital), with four pathologists and 2 private pathology labs. The registrar abstracts details of all cancer diagnoses (address is usually unavailable). Haematological malignancies are all diagnosed by the registry director and his junior- Dr. Lotodo (Department of Haematology and Blood Transfusion).

Death registration is virtually universal (home deaths should also be autopsied, but this is a rare event). The death certificate may not be issued for some weeks after death, however. Deaths at home are certified by the village chief, based on relatives’ assessments, and are of very questionable accuracy (often just “cancer”).

Cases are coded in the registry, and data entry and management is by CANREG-5. By January 2013, some 16,000 cases had been registered, about one half of which are in residents of Uasin Gishu district.

3 Study Population

Eldoret: Situated some 330 km northwest of Nairobi, Eldoret is the largest city and the administrative capital of Uasin Gishu County.

Moi’s Bridge: Located some 40 km north of Eldoret along Eldoret – Kitale road, Moi’s bridge is a small yet an important agricultural town.

Burnt Forest: Sited some 35 kilometers to Eldoret, Burnt Forest is the first major town as you approach Eldoret from Nairobi.

Turbo: Turbo is a small agricultural town located about 31 km west of Eldoret town along Eldoret -Webuye road.

Others: Outside the county

drawing

4 Objective

The main objective of this study/analysis is to assess the incidence and prevalence of different cancer types from 2000 to 2020

5 Results for 2001 to 2005

5.1 Data Quality Assessment

5.1.1 DQA Summary Tables

Data summary
Name clean.df
Number of rows 77271
Number of columns 50
_______________________
Column type frequency:
character 7
factor 22
logical 1
numeric 20
________________________
Group variables None

Variable type: character

skim_variable n_missing complete_rate min max empty n_unique whitespace
SOURCE2 65159 0.16 1 2 0 68 0
SOURCE3 75280 0.03 1 2 0 46 0
SOURCE4 76898 0.00 2 2 0 23 0
PRIMARY_ 762 0.99 2 3 0 242 0
ICD10 12702 0.84 3 4 0 337 0
ICCC 18351 0.76 1 3 0 79 0
DATEB 1055 0.99 1 8 0 3508 0

Variable type: factor

skim_variable n_missing complete_rate ordered n_unique top_counts
SOURCE1 128 1.00 FALSE 84 MR&: 29327, HIS: 12700, DEA: 5128, HAE: 3650
CHECKS 0 1.00 FALSE 4 Don: 63898, Not: 12844, Don: 483, Don: 46
ADRCODE 0 1.00 FALSE 8 Unk: 62687, Soy: 2662, TUR: 2574, KAP: 2208
T 71540 0.07 FALSE 8 T3: 1765, T4: 1582, T2: 1417, T1: 488
M 71601 0.07 FALSE 4 M0: 3188, M+: 1704, M1: 444, #!N: 334
TRIBE 741 0.99 FALSE 48 Kal: 20583, LUH: 14740, LUO: 9674, Nan: 7898
Sex 1 1.00 FALSE 2 Fem: 40776, Mal: 36494
Morphology 70 1.00 FALSE 317 Squ: 13504, Ade: 9154, Inf: 6305, Neo: 6188
Behaviour 42 1.00 FALSE 4 Mal: 77131, In : 54, Unc: 40, Ben: 4
Basis of Diagnosis 64 1.00 FALSE 10 His: 59423, Cyt: 10434, Cli: 4147, Dea: 1613
Stage 29624 0.62 FALSE 7 Unk: 26632, Sta: 11193, Sta: 5833, Sta: 2606
Grade 6766 0.91 FALSE 9 Unk: 48680, Gra: 9860, Gra: 7040, Gra: 4387
Surgery 3990 0.95 FALSE 3 No: 42280, Unk: 19053, Yes: 11948
Radio Therapy 4033 0.95 FALSE 3 No: 44575, Unk: 20113, Yes: 8550
Chemo Therapy 3915 0.95 FALSE 4 No: 31084, Yes: 29672, Unk: 12595, #!N: 5
Hormonal Therapy 4045 0.95 FALSE 3 No: 47828, Unk: 21311, Yes: 4087
Symptoms 202 1.00 FALSE 4 Yes: 76447, No: 445, Unk: 175, #!N: 2
Mortality Status 1020 0.99 FALSE 2 Ali: 55497, Dea: 20754
Cause of Death 72794 0.06 FALSE 3 Can: 4381, Unk: 59, Oth: 37
Concurrent Illness 1305 0.98 FALSE 4 Unk: 57864, ABS: 11730, PRE: 6371, #!N: 1
Primary Site 0 1.00 FALSE 60 BLO: 8640, BRE: 7928, CER: 7779, ESO: 6719
Age Group 786 0.99 FALSE 16 60-: 6983, 55-: 6937, 40-: 6747, 50-: 6576

Variable type: logical

skim_variable n_missing complete_rate mean count
MPCODE 77271 0 NaN :

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
SOURCERECORDID 21 1.00 2.014457e+13 5.544472e+10 1.987000e+13 2.012104e+13 2.016168e+13 2.018341e+13 2.021182e+13 ▁▁▁▃▇
AGE 0 1.00 4.682000e+01 2.285000e+01 -1.000000e+00 3.300000e+01 4.900000e+01 6.300000e+01 9.990000e+02 ▇▁▁▁▁
INCID 203 1.00 2.014201e+07 1.812936e+05 2.010110e+06 2.012082e+07 2.016073e+07 2.018092e+07 2.020122e+07 ▁▁▁▁▇
HISTO 70 1.00 8.588660e+03 6.653400e+02 8.000000e+03 8.070000e+03 8.140000e+03 9.140000e+03 9.989000e+03 ▇▂▂▁▂
BEHA 42 1.00 3.000000e+00 6.000000e-02 0.000000e+00 3.000000e+00 3.000000e+00 3.000000e+00 3.000000e+00 ▁▁▁▁▇
GRADE 6766 0.91 6.890000e+00 3.190000e+00 1.000000e+00 3.000000e+00 9.000000e+00 9.000000e+00 9.000000e+00 ▂▁▁▁▇
BASIS 64 1.00 6.250000e+00 1.660000e+00 0.000000e+00 7.000000e+00 7.000000e+00 7.000000e+00 9.000000e+00 ▁▁▂▇▁
STAGE 29624 0.62 6.480000e+00 2.910000e+00 0.000000e+00 4.000000e+00 9.000000e+00 9.000000e+00 9.000000e+00 ▁▂▃▁▇
SURG 3990 0.95 3.240000e+00 3.430000e+00 1.000000e+00 1.000000e+00 1.000000e+00 9.000000e+00 9.000000e+00 ▇▁▁▁▃
RADIO 4033 0.95 3.310000e+00 3.510000e+00 1.000000e+00 1.000000e+00 1.000000e+00 9.000000e+00 9.000000e+00 ▇▁▁▁▃
CHEMO 3915 0.95 2.780000e+00 2.870000e+00 1.000000e+00 1.000000e+00 2.000000e+00 2.000000e+00 9.000000e+00 ▇▁▁▁▂
HORMONE 4045 0.95 3.380000e+00 3.610000e+00 1.000000e+00 1.000000e+00 1.000000e+00 9.000000e+00 9.000000e+00 ▇▁▁▁▃
SYMPT 202 1.00 2.010000e+00 3.400000e-01 0.000000e+00 2.000000e+00 2.000000e+00 2.000000e+00 9.000000e+00 ▁▇▁▁▁
REGNO_ 21 1.00 2.014433e+07 8.563807e+04 2.002853e+06 2.012104e+07 2.016168e+07 2.018341e+07 2.021182e+07 ▁▁▁▁▇
SEX 1 1.00 1.530000e+00 5.000000e-01 1.000000e+00 1.000000e+00 2.000000e+00 2.000000e+00 2.000000e+00 ▇▁▁▁▇
CONC_ILL 1305 0.98 7.250000e+00 3.140000e+00 1.000000e+00 9.000000e+00 9.000000e+00 9.000000e+00 9.000000e+00 ▂▁▁▁▇
DATELAST 4064 0.95 2.014461e+07 5.475913e+05 0.000000e+00 2.013022e+07 2.017051e+07 2.019050e+07 1.000000e+08 ▁▇▁▁▁
STATUS 1020 0.99 1.270000e+00 4.500000e-01 1.000000e+00 1.000000e+00 1.000000e+00 2.000000e+00 2.000000e+00 ▇▁▁▁▃
CAUSE 72794 0.06 1.110000e+00 9.200000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 9.000000e+00 ▇▁▁▁▁
PATIENTRECORDID 21 1.00 2.014433e+09 8.563798e+06 2.002853e+08 2.012104e+09 2.016168e+09 2.018341e+09 2.021182e+09 ▁▁▁▁▇

5.1.2 DQA Report: Main Issues

  • This can’t be right? 77K casese between 2001 and 2005?
  • Date columns doesn’t look right, was it exported using different format?
  • Missing Dictionary for: ADRCODE: “041” “047” “052” “042” “001” “010” “032” “040” "070
  • Missing Dictionary for: HIST: “8692” “8821” “9766” “8822” “8077” “8472” “9340” “9537” “9975” “9300” “9970” “9421” “9352” “8507” “8870”
  • Missing Dictionary for: STAGE: 5, 7
  • Missing Dictionary for: CONC_ILL: 4
  • Missing Dictionary for: T: “+”
  • Missing Dictionary for: M: 9,4,2,3,X
  • Missing Dictionary for: SOURCE1: BV
  • Missing Dictionary for: TRIBE: 7, 48, 45, 31

5.2 Study Sample Analysis

5.2.1 Plots

5.2.2 Socio-Demographic Characteristics

5.2.3 Sub County distribution

5.3 General Summary Statistics

5.3.1 Main Covariates

5.3.2 Primary Cancer Site

5.3.3 Morphology

5.4 Main Analysis

5.4.1 Plot: Frequency and Proportions of the Most Common Cancer Types

5.4.2 Summary Table: : Frequency and Proportions of the Most Common Cancer Types

5.5 Mortality Analysis

5.5.1 Mortality Status

5.5.1.1 Plot

5.5.1.2 Summary Table | Age

5.5.2 Mortality Risk Factors

5.6 Clinical Factors Analysis

5.6.1 Clinical Summary

5.6.1.1 Plot: Cancer characteristics

5.6.1.2 Plot: Treatment

5.6.1.3 Summary table

5.6.2 Basis of Diagnosis Distribution

5.6.2.1 Plot

5.6.2.2 Summary Table

5.6.3 Behaviour Distribution

5.6.3.1 Plot

5.6.3.2 Summary Table

5.6.4 Staging Distribution

5.6.4.1 Plot

5.6.4.2 Summary Table

5.6.5 Grade Distribution

5.6.5.1 Plot

5.6.5.2 Summary Table

6 Results for 2006 to 2010

TODO

7 Results for 2011 to 2015

TODO

8 Results for 2016 to 2021

TODO

9 Results Overall | 2001 to 2021

TODO

10 Appendix

10.1