Nigerian Journal of Paediatrics

Relationship of socio economic status and childhood cancer an in hospital cross sectional study in a developing country

Niger J Paediatr 2017; 44 (4): 180 – 184

ORIGINAL

Brown BJ

CC – BY Relationship of socio-economic

Adeleye AO

status and childhood cancer: an

in-hospital cross-sectional study in

a developing country

DOI:http://dx.doi.org/10.4314/njp.v44i4.2

Accepted: 30th November 2017

Abstract : Background: Socio-

cancer but only 7 (8.1%) knew it

economic factors are known to

could occur in children. There was

Brown BJ (

)

affect health quality, disease oc-

no association between Burkitt

Adeleye AO

currence as well as health-seeking

lymphoma and socio-economic

Department of Paediatrics,

behaviors in several ways.

class. Twenty-eight (30.8%) par-

Department of Surgery,

Objectives: To determine the in-

ents of the 91 children visited al-

University College Hospital,

fluence of socio-economic factors

ternate sources of health care, most

Ibadan, Nigeria.

on awareness of cancer, health-

commonly traditional healers, fol-

Email: biosbrown@yahoo.com

seeking behaviors among parents

lowed by religious centers. There

of children with cancer in a devel-

was no association between visits

oping country and occurrence of

to such centers and the parents’

cancer using Burkitt lymphoma as

socio-economic status or with

index malignancy.

presentation with metastatic dis-

Methods: This was a descriptive

ease.

cross-sectional study that in-

Conclusions: Awareness of child-

volved children with cancer seen

hood cancer is low among this

over a 2-year period in a tertiary

cohort of parents; their socio-

hospital in Nigeria. Information

economic status seems to impact

was

obtained

interview

on this level of awareness but not

through administration of a ques-

on their health-seeking behaviors

tionnaire and retrieval of clinical

for their affected children.

Fo-

data from patients’ case notes.

cused health education is needed to

Results: The caregivers of 91 chil-

increase childhood cancer aware-

dren (46 boys, 45 girls) were in-

ness and

appropriate health-

terviewed including 86 biological

seeking behavior among the popu-

parents. Majority (84.6%) of the

lation studied.

children belonged to the low socio

-economic classes 3-5; 45 of 86

Key

words :

socio-economic;

parents (52.3%), more likely in

childhood; cancer; health-seeking;

parents

from

higher

socio-

behaviour; awareness

economic classes, were aware of

Introduction

accessing of clinical services aimed at early diagnosis of

cancer. Knowledge of early symptoms or signs of can-

Social and economic factors are known to affect health

cer is believed to be important in facilitating early help-

and well-being in several ways. For example, individu-

seeking from health facilities and consequently, early

diagnosis. However, knowledge of early clinical fea-

als of lower socio-economic status (SES) have poorer

health metrics than those in higher socio-economic

tures can only be predicated, in the first instance, on

classes. Similarly, children with lower SES have been

awareness of the existence of cancer.

reported to have poorer health indices compared with

those in the higher SES.

By the same token, socio eco-

In another light, socio-economic factors have also been

nomic and cultural factors are believed to influence help

considered to play a role on the occurrence of cancer.

-seeking attitudes in childhood cancer; this usually re-

For example, the risk of cancers of the lungs and stom-

sults in late in-hospital presentation and subsequent poor

ach are higher in socially disadvantaged groups while

outcomes in Nigeria.

3,4

The influence of cultural factors

that of cancer of the breast is higher in socially advan-

taged groups

on health access are not limited to developing countries.

More specifically, descriptive studies

A study in the United Kingdom revealed that blacks and

have shown that most children with Burkitt lymphoma,

ethnic minorities had poor knowledge of cancer and also

the most common childhood tumour in Nigeria, belong

to the low socio-economic classes.

8,9

had beliefs and attitudes that might hinder their

Due to lack of ana-

181

lytical studies on the subject, it is not clear if this pattern

Table 1: Parameters used in classification of socio-economic

reflects the low socio-economic status of the general

status

populace; whether this pattern is peculiar to Burkitt

Parameter

Lymphoma in contradistinction to other tumours, or if

Educational level

there is a statistical association between Burkitt Lym-

University graduates or equivalents

phoma and socioeconomic class. The latter may indicate

School certificate (SSCE/GCE O’Level) holders who also

some contribution of socio-economic class to the occur-

have teaching or other professional training

rence of the disease and therefore a potential target for

School certificate or grade II teachers’ certificate Holders

control measures. The objectives of this study were

or equivalents

therefore, to determine the influence of socio-economic

Modern 3 and primary six certificate holders

factors on awareness of cancer among parents of a co-

Those who can just read and write or the illiterate

hort of children with cancer, help-seeking by these par-

ents from alternative medicine and to compare the socio-

Occupation

economic status in children with Burkitt lymphoma as a

Senior public servants, professionals, managers,

reference with those of other childhood cancers. Confir-

Large scale traders, businesswomen and contractors

mation of the significance of these factors would give

Intermediate grade public servants, senior schoolteachers

impetus to educational drives and other efforts aimed at

Junior schoolteachers, drivers and artisans

their remediation.

Petty traders, messengers, labourers and similar grades

Unemployed, fulltime housewives, students and Subsistence

farmers

Materials and Methods

Information on the pathologic diagnosis and stage of the

disease was obtained from the case notes.

This was a cross sectional descriptive study involving all

Ethical approval was obtained from the Joint University

children diagnosed with cancer at the Pediatric Hematol-

of Ibadan/University College Hospital, Ibadan Ethics

ogy and Oncology Unit of the University College Hospi-

Committee and informed consent was obtained from the

tal, Ibadan, South-western Nigeria. This hospital is an

parents or guardians.

850-bed tertiary health facility that takes care of both

adults and children. It serves as a referral center for can-

Data were entered into a micro-computer using the Sta-

cer in Oyo state in which it is located, as well as some

tistical Package for Social Sciences, version 22.0. Means

surrounding cities and states of the nation. Nigeria has a

and medians were computed and categorical variable

dual health service consisting of private and public ser-

presented in frequencies and proportions. Association

vices. The hospital in which the present study was car-

between categorical variables was tested using the Chi-

ried out is a public hospital. The health structure of the

Square test and when applicable, the Fisher’s Exact test.

country comprises three levels of care namely primary,

Non-parametric variables were compared using the In-

secondary and tertiary.

Childhood cancer is managed

dependent samples Median test. Statistical significance

mainly at government owned tertiary centers, so that

was set at p < 0.05.

children irrespective of their social classes are likely to

be managed mainly at tertiary health centers. There is a

National Health Insurance Scheme which is utilized by

only about 4 percent of the population. Therefore, pay-

Results

ment for health services in the country is mainly through

out-of-pocket expenditure.

11, 12

The caregivers of 91 children with cancer comprising 46

boys and 45 girls were interviewed and questionnaires

The present study took place over a period of two years

completed. Out of the 91 respondents 17 (18.7%) were

from July 2012-June 2014. Information was obtained by

fathers, 69 (75.8%) mothers and the remaining 5 (5.5%)

interview through administration of a questionnaire and

were 3 grandmothers, 1 uncle and 1 aunt. The diagnoses

retrieval of clinical data from the patients’ case notes.

in the children are shown in table 2 with Retinoblastoma

Information obtained through interview included socio-

accounting for the majority.

demographic data, awareness of cancer, and health-

The distribution of the children across the 5 socio-

seeking from alternative sources of health care. Stratifi-

economic classes was as follows: 4 (4.4%) children in

cation of the children into socio-economic classes was

class 1; 10 (11.0%) in class 2; 49 (53.8%) in 3; 27

done with the classification by Oyedeji which is based

(29.7%) in 4, and only 1 (1.1%) child in class 5.

on the parent’s level of education and occupation. The

Analysis of the relationship between socio-economic

respective strata of this classification system are shown

status of children and tumors was done using Burkitt

in table 1. The mean of four scores (one each from the

Lymphoma as index malignancy. Comparing the socio-

father’s educational level and occupation and also from

economic status of families of children with Burkitt

the mother’s educational level and occupation) to the

Lymphoma with other children revealed that none (0%)

nearest whole number was the social class assigned to

of the 12 children with Burkitt Lymphoma belonged to

the child on a 5-point scale

with class 1 representing

the higher socio-economic classes (1&2); 12 (100%)

the highest level of socio-economic status, and 5 the

belonged to classes 3-5, whereas 14 (17.7%) of the 79

lowest.

children with other malignancies belonged to the high

socio-economic classes 1&2, the rest 65 (82.3%) be-

182

longing in the lower socio-economic classes 3-5. This

gious faith healers by 15(16.5%) and Islamic religious

difference was however not statistically significant

healers by 15(16.5%) of study participants. Time spent

(Fisher’s Exact test, p = 0.201).

in such places receiving treatment ranged from 1-15

days with a median of 2 days. The median interval be-

Among the 86 parents, 45 (52. 3%) had prior awareness

tween onset of symptoms and diagnosis was 18.2 weeks

of cancer as an ailment but only 7 (8.1%) knew cancer

among children who visited alternative sources of health

could occur in children. Regarding knowledge about

and 13.0 weeks in those who did not visit this route

causation of disease, 23 (26.7%) of the 86 parents be-

(Independent samples median test p =0.228).

lieved that the source of their children’s illness was

Metastasis was present in 7 (50%) of the 14 children

spiritual; 47 (54.7%) were not sure of the cause of the

from higher socio-economic classes 1&2 compared with

illness, while only 16 (18.6%) believed it to be physical.

33 (42.9%) of the 77 patients of lower socio-economic

Awareness of cancer both in general and its occurrence

classes (Chi-Square test, p =0.620). Metastatic disease

in childhood was significantly associated with a higher

was present in a higher proportion, 16 (57.1%) of 28

socio-economic class (Table 3).

children who visited alternative sources of health com-

pared with only 24 (38.1%) of 63 who never took this

Table 2: Types of malignancy found in the study population

health-access route but the difference was not statisti-

Diagnosis

Frequency

Percent

cally significant (Pearson’s Chi square test, p= 0.091).

Burkitt Lymphoma

13.2

Other Non-Hodgkin Lymphomas

5.5

Hodgkin Lymphoma

1.1

Retinoblastoma

26.4

Discussion

Rhabdomysarcoma

14.3

Nephroblastoma

9.9

The findings concerning the main theme of the study

Neuroblastoma

2.2

suggest a notable trend of high patronage of comple-

Central Nervous System tumor

9.9

mentary alternative medicine by this study group. This

Leukemia

11.0

trend has been reported in high-income countries too.

Bone tumors

2.2

Hepatoblastoma

2.2

Almost a third of the children in this study at some point

Testicular germ cell tumor

1.1

in time patronized alternative sources of health care such

Parotid tumor

1.1

as churches, Islamic mission houses, as well as tradi-

Total

100.0

tional/herbal medicine practitioners. This proportion is

lower than the rate in a Swiss University Hospital where

53 percent of respondents had used Complementary and

Table 3: Relationship between parental socio-economic class

Alternative Medicine (CAM) for their children with

(SEC) and cancer awareness and presumed cause

cancer, mostly homeopathy.

SEC 1-2

SEC 3-5

+p-value

(n=13)

(n=73)

Importantly however, this study showed that there was

Frequency

(%)

no statistically significant association between socio-

economic status and the use of alternative sources of

Cancer aware-

Yes

11(84.6)

34 (46.6%)

0.015

ness

health care by the subjects. This suggests that this health

2 (18.2)

39 (53.4)

-seeking behavior knows no socio-economic barriers but

Childhood cancer

Yes

4 (30.8)

3 (4.1)

0.009

is more probably a reflection of the culture of the peo-

awareness

ple. Health educational efforts to prevent such practices

9 (69.2)

70 (95.9)

should therefore cut across all strata of the society. It is

Cause of illness

Spiritual

0.499

also noteworthy that patronizing such alternative sources

Non-

spiritual

of health services did not significantly contribute to de-

layed diagnosis of cancer. Although the pre-diagnostic

interval was longer in children who patronized alterna-

+Fisher’s exact test

tive health care sources, the difference was not statisti-

cally significant. Similarly, presentation with metastatic

Attribution of the cause of cancer to spiritual factors was

disease was not significantly higher in children who had

not associated with socio-economic class (Table 3). Me-

used alternative medicine. This is not surprising, giving

dian time between onset and diagnosis in children whose

the very short median time (only 2 days) spent in such

parents thought the illness was of spiritual origin was

places. This contrasts with findings in breast cancer pa-

25.8 weeks compared to 33.1 weeks in others

tients both in Nigeria and Pakistan where the use of al-

(Independent samples median test, p =0.896).

ternative and complementary medicine (CAM) signifi-

Twenty-eight (30.8%) of the 91 children had been taken

cantly has been shown to increase pre-diagnostic delay

to alternative sources of health care before in-hospital

and presentation with advanced disease.

16,17

This differ-

presentation.

Out of the families from high socio-

ence in the impact of using CAM in children in this

economic classes 1&2, 21.4% (3/14) visited alternative

study compared to adult breast cancer patients may be

sources of health care compared to 25 of the 77 (32.5%)

due to fears of early death of children due to their vul-

from lower socio-economic classes (Fisher exact test, p=

nerability and therefore minimal length of time spent in

0.537). The alternative medicine sources visited were

CAM. Therefore rather than lay all the blame on the

traditional herbal healers by 17(18.7%), Christian reli-

183

parents and visits to alternative medical practitioners for

Burkitt lymphoma in Nigeria have revealed that most

children were from low socio-economic classes. How-

delayed diagnosis in children, health system defects

should be re-appraised. The need for this is corroborated

ever, a statistical association between the condition and

by a recent report that revealed how inadequacies of

socio-economic class has neither been tested for, nor

health systems might be contributory to delayed diagno-

established. The uniqueness of the present study, unlike

sis of childhood cancer. Some of these factors include

previous ones is that it compared socio-economic

late referrals to tertiary centers, a chain of multiple refer-

classes of children with Burkitt Lymphoma with those

rals before arriving at the place where diagnosis is made

of children with other tumours. Although the present

and out-of-pocket financing of cancer treatment.

study, like the ones before, reveals a predominance of

low socio-economic classes in affected children as a

Delay in the interval between onset of symptoms of can-

whole, no statistically significant association with socio-

cer and diagnosis or treatment has been associated with

economic status was found. The established fact of the

a poorer survival. Lack of awareness of the seriousness

predominance of children of low socio-economic class

of symptoms or not recognizing the symptom to be due

in Burkitt Lymphoma may therefore reflect nothing

to cancer is a major risk factor for delayed presentation

more than the picture in the general population of chil-

of patients. In the present study, only 8.1% of parents

dren with tumours since only 15.4 % of the study popu-

were aware that cancer could occur in children. This

lation belonged to the high socio-economic classes.

finding is in keeping with the low awareness rates also

Indeed, it may also actually just be a reflection of the

reported in guardians of 5.0% in Kenya and 19% in

distribution of socio-economic classes across the general

Uganda.

This implies that the index of suspicion of

population but the validation of this impression is be-

cancer is likely to be low when children develop symp-

yond the scope of this present work.

toms of the disease and so foster delayed health-seeking

in appropriate facilities. The association between higher

socio-economic class and cancer awareness in this study

is in keeping with findings by other workers

. 21

Our find-

Conclusions

ings highlight the need for establishment of childhood

cancer awareness programmes, paying extra attention to

This study has highlighted a low level of awareness of

individuals from lower socio-economic classes. This is

occurrence of cancer in childhood among mothers of

vital since cancer awareness is a potentially modifiable

affected children in this developing country, the health-

contributor to the variations seen in healthcare seeking

seeking behaviour of parents of children with cancer

and, ultimately, survival.

The present study also did

with a high rate of pre-hospital consultation of CAM,

not show any association between socio-economic status

and some socio-economic and cultural issues in child-

and presentation with metastatic disease. This is in con-

hood cancer. Socio-economic status is associated with

trast with findings in a study on Osteosarcoma in the

awareness of cancer but not with perception of its causa-

United States where individuals from low socio-

tion, utilization of alternative medicine and presentation

economic classes were more likely to present with me-

with metastatic disease within the study population. This

tastatic disease.

The contrasting finding of this study

study has also confirmed a lack of statistical association

concerning this point may be due to other overriding

between Burkitt lymphoma and socio-economic class.

cultural factors or barriers in accessing health that per-

Health educational efforts and health system reforms

vade the entire socio-economic landscape of the health

aimed at promoting early diagnosis and better outcomes

system in Nigeria.

12, 24

for childhood cancer are recommended and should cut

across all strata of our society.

Although Burkitt Lymphoma is the most common child-

hood malignant tumour in Nigeria, its ranking second in

the present study is in keeping with declining frequency

Conflicts of interest: None

of the tumour in Ibadan, which has been attributed to

Funding: None

possibility of improved malaria control. Studies on

References

1. Chen E. Why Socioeconomic

Meremikwu MM, Ehiri JE,

Thomas VN, Saleem T, Abra-

Status Affects the Health of

Nkanga DG, Udoh EE, Ikpatt

ham R. Barriers to effective

Children A Psychosocial Per-

OF, Alaje EO. Socioeconomic

uptake of cancer screening

spective. Curr Dir Psychol Sci

constraints to effective man-

among Black and minority

2004; 13:112 – 115.

agement of Burkitt’s lym-

ethnic groups. Int J Palliat

Currie A, Shields MA, Price

phoma in south-eastern Nige-

Nurs 2005; 11:562, 564 – 571.

SW. The child health/family

ria. Trop Med Int Health 2005;

Robb K, Stubbings S, Rami-

income gradient: Evidence

10:92 – 98.

rez A, Macleod U, Austoker J,

from England. J Health Econ

Brown BJ, Adeleye AO, Ibeh JN.

Waller J, et al . Public aware-

2007; 26:213 – 232.

A Prospective Study on the Causes

ness of cancer in Britain: a

of Delayed Diagnosis of Child-

population-based survey of

hood Cancer in Ibadan, Nigeria.

adults. Br J Cancer 2009; 101

Pediatr Hematol Oncol 2015;

(Suppl 2):S18 – 23.

32:365 – 373.

184

Faggiano F, Partanen T, Koge-

15. Magi T, Kuehni CE, Torchetti

21. Dey S, Mishra A, Govil J,

vinas M, Boffetta P: Socioeco-

L, Wengenroth L, Lüer S, Frei-

Dhillon PK. Breast Cancer

nomic differences in cancer

Erb M. Use of Complementary

Awareness at the Community

incidence and mortality. In:

and Alternative Medicine in

Level among Women in

Kogevinas M, Pearce N, Susser

Children with Cancer: A Study

Delhi, India. Asian Pac J Can-

M, Boffetta P, editors. Social

at a Swiss University Hospital.

cer Prev 2015; 16:5243-51.

inequalities and cancer. Volume

PLoS One 2015; 10

22. Richards MA. The National

138. Lyon: IARC, IARC Scien-

(12):e0145787.

Awareness and Early Diagno-

tific Publications; 1997:65-

16. Ezeome ER. Delays in presen-

sis Initiative in England: as-

176.

tation and treatment of breast

sembling the evidence. Br J

Brown B. A review of the lit-

cancer in Enugu, Nigeria. Niger

Cancer. 2009; 101(Suppl

erature on childhood Burkitt

J Clin Pract 2010; 13:311 –

2):S1 – S4.

lymphoma in Nigeria. Niger J

316 .

23. Miller BJ, Cram P, Lynch CF,

Paediatr 2016; 43:1-7.

17. Malik IA, Gopalan S. Use of

Buckwalter JA. Risk factors

Oguonu T, Emodi I, Kaine W.

CAM results in delay in seek-

for metastatic disease at pres-

Epidemiology of Burkitt’s lym-

ing medical advice for breast

entation with osteosarcoma:

phoma in Enugu, Nigeria. Ann

cancer. Eur J Epidemiol 2003;

an analysis of the SEER data-

Trop Paediatr. 2002; 22:369 –

18:817 – 822.

base. J Bone Joint Surg Am

74.

18. Richards MA, Westcombe AM,

2013; 95:e89.

10. Welcome MO. The Nigerian

Love SB, Littlejohns P, Rami-

24. Adeleye AO, Joel-Medewase

health care system: Need for

rez AJ. Influence of delay on

VI. Awareness and uptake of

integrating adequate medical

survival in patients with breast

measures for preventing CNS

intelligence and surveillance

cancer: a systematic review.

birth defects among mothers

systems. J Pharm Bioall Sci

Lancet (London, England)

of affected children in a sub-

2011; 3:470-478.

1999; 353(9159):1119 – 1126.

Saharan African neurosur-

11. Onoka CA, Onwujekwe OE,

19. Ramirez AJ, Westcombe AM,

geon’s practice. Child’s Nerv

Uzochukwu BS, Ezumah NN.

Burgess CC, Sutton S, Lit-

Syst 2015; 31: 2311 – 2317.

Promoting universal financial

tlejohns P, Richards MA. Fac-

25. Babatunde TO, Akang EEU,

protection: constraints and ena-

tors predicting delayed presen-

Ogun GO, Brown BJ. Pattern

bling factors in scaling-up cov-

tation of symptomatic breast

of childhood cancer in Univer-

erage with social health insur-

cancer: a systematic review.

sity College Hospital, Ibadan

ance in Nigeria. Health Res

Lancet (London, England)

during 1991-2010 and com-

Policy Syst 2013; 11: 20.

1999; 353:1127 – 1131.

parison with the previous

12.

Adeleye AO, Jite IE, Smith

20. Buckle GC, Collins JP, Sumba

three decades. Paediatrics and

OA. A tale of two acute ex-

PO, et al . Factors influencing

International Child Health

tradural hematomas. Surg Neu-

time to diagnosis and initiation

2015; 35: 144 – 50.

rol Int 2016; 7:54.

of treatment of endemic Burkitt

13. Oyedeji G. Socio-economic

Lymphoma among children in

and Cultural Background of

Uganda and western Kenya: a

Hospitalised Children in Ilesha.

cross-sectional survey. Infect

Niger J Paediatr 1985; 12:111

Agent Cancer 2013; 8:36.

– 117.

14. Horneber M, Bueschel G, Den-

nert G, Less D, Ritter E,

Zwahlen M. How Many Cancer

Patients Use Complementary

and Alternative Medicine: A

Systematic Review and Meta-

analysis. Integr Cancer Ther

2012; 11:187 – 203.