Identifikasi Variabel Confounding dengan Penerapan Uji Chi Square Mantel Haenzel Pada Hubungan Antenatal Care (ANC) Terhadap BBLR di Kota Samarinda

Abstract

Background: Confounding is a very important issue to consider because its presence can affect the p-value and the magnitude of the risk that can cause errors in decision making. The Mantel Haenszel is used when there are one or more confounding variables, which can be applied to public health problems such as the incidence of LBW which has many risk factors such as antenatal care (ANC), maternal age, parity, and anaemia.

Objective: To identify confounding variables in the ANC relationship with LBW.

Method: Nonreactive research with a case-control design. The sample was 305 (61 cases; 244 controls), in 3 Samarinda City Health Centers and sourced from the 2016-2017 maternal cohort registered, East Kalimantan.

Results: Maternal age was confounding in the relationship of ANC to LBW (ΛOR 12.29%) whereas parity (ΛOR 1.65%) and anaemia (ΛOR 0,64%) didn’t become confounding.

Conclusion: ANC has a significant relationship with the incidence of LBW without or accompanied by confounding variables. Parity and anemia are not as confounding whereas maternal age is influential confounding which is marked by changes in OR before and after considering external variables. Thus, more intensive pregnancy care is needed more mothers who are it too young and or too old.