Limitations and recommendations for use of secondary data analysis in pediatric research

Fleischer, Nicole J.; Khalil, Amani

doi:10.1080/02739615.2023.2279064

Cited by 2 publications

(1 citation statement)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the use of a secondary dataset from a cross-sectional survey exposed the study to some specific limitations [ 7 , 22 ]. We may not have controlled all covariates since those needed for this study may not have been collected, also, some important variables may have been deleted to conceal identity of participants; hence, not available for analysis [ 35 , 36 ]. In addition, we recoded categorical and continuous variables for this research, such action according to studies [ 37 ] has the potential to generate an insufficient or biased estimate.…”

Section: Discussionmentioning

confidence: 99%

Adolescent pregnancy persists in Nigeria: Does household heads’ age matter?

Mbulu,

Yang,

Wallen

2024

PLOS Glob Public Health

View full text Add to dashboard Cite

About 700,000 pregnant youths die each year in developing countries. To determine whether the persistent adolescent pregnancy in Nigeria between 2013 and 2018 was influenced by proximal factors, particularly household head age, we carried out a cross-sectional study on adolescent girls that participated in the 2018 Nigeria Demographic and Health Survey (NDHS). Age of first birth, residence type, age, and gender of household heads was collected using a 2018 standardized NDHS. Multiple logistic regression was performed to test for associations. We analyzed 8,448 adolescents who had experienced pregnancy during the study period. The results demonstrated that girls with male household heads aged 45 and older have lower odds of adolescent pregnancy (OR = 0.619; 95% CI = 0.447, 0.856; p = 0.004 compared to those with female household heads in the same age group. Girls with male household heads in three younger age groups have higher odds of adolescent pregnancy than those with female household heads in the corresponding age group (15–24: OR = 1.719, 95% CI = 1.042, 2.835, p = 0.034; 25–34: OR = 4.790, 95% CI = 1.986, 11.551, p < 0.001; 35–44: OR = 2.080, 95% CI = 1.302, 3.323, p = 0.002). Girls with household heads aged in the 15–24 and 25–34 groups had higher odds of adolescent pregnancy compared to those with household heads aged 45 and older. Higher odds of adolescent pregnancy in Nigeria were found among girls with household heads aged in the 15–24 and 25–34 groups compared to those with household heads aged 45 and older. Although girls with younger male household heads are at an increased risk nationally, those living in the rural areas with younger household heads are at an even higher risk for adolescent pregnancy. Therefore, levels of socioecological model must be considered in planning for effective interventions.

show abstract