Abstract
Interventions and schemes aimed at reducing the risks associated with pregnancy are often implemented wholesale. This often leads to misdirection of interventions to the wrong patients. To address this, there was the need to extract segments that subgroup maternal attributes into frequently occurring patterns. Some secondary data consisting of records of pregnant women who attended antenatal care (ANC) visits at a hospital were subjected to association rules mining. The analysis extracted and sub-grouped the attributes and characteristics of pregnant women that often co-occur. Segmentation was done in three sub-groups. Each segment consists of both positive and risk attributes/characteristics that often occur together in a pregnant woman. With the aid of the segments, intervention programs can be designed and delivered according to the needs and requirements of each segment. This provokes a new perspective on how quality healthcare service delivery can be channeled to pregnant women based on specific needs.
TopIntroduction
Segmentation is an activity that subgroups individuals or objects in a particular domain into segments (clusters) that share a similarity or pattern. Customer segmentation, for instance, allows business enterprises to develop strategies and products that target the specificities of each subgroup (Sari, Nugroho, Ferdiana, & Santosa, 2011). In the context of pregnancy, segmentation can be seen as the process of subdividing pregnant women into subgroups based on similar attributes, characteristics, and antecedents from existing data.
There are several intervention and support programs across the world which are targeted at pregnant women to ensure positive pregnancy outcomes. A study by Rutman, Hubberstey, Poole, Schmidt, and Bibber (2020) identified eight different programs in Canada designed to provide support for pregnant women. The programs are structured to offer basic needs, primary, mental, and perinatal healthcare to pregnant women. Furthermore, substance use, trauma, and violence services are offered to pregnant and parenting women through these programs (Rutman et al., 2020). Similarly, the World Health Organization (WHO) reported a program designed to provide pregnancy, postnatal, and baby care support information to women in South Africa (WHO, 2013). The program, termed MAMA SMS, enables pregnant and nursing women to enroll for a free messaging service. Through this platform, support information is disseminated regularly on themes such as HIV testing, clinic attendance, mother-infant bonding, preparing for birth, hygiene, adherence to medication, nutrition, prevention of sexually transmitted infections (STIs), and so on. Part of the objectives of this service is to ensure that women remain healthy throughout the pregnancy duration, encourage HIV testing and adherence to the guidelines on prevention of mother-to-child transmission (PMTCT) of HIV (WHO, 2013). A study by Feijen-de Jong, Warmelink, van der Stouwe, Dalmaijer, and Jansen (2022) identified several intervention programs for pregnant women in the North of the Netherlands. These include Nurse Family Partnership (NFP), Supportive Parenting (SP), Mothers Inform Mothers (MIM), and Meeting Centre for Young Parents (MCYP). The NFP program assigns nurses to visit vulnerable pregnant women at home every month to provide empowerment and support during pregnancy. The visits continue until the child is delivered and attains two years. The SP program focuses on providing parenting skills and support to pregnant women up to the first 18 months after delivery. A health professional is assigned to the pregnant woman to make a total of six home visits during the intervention. On the other hand, the MIM program co-opts experienced mothers who volunteer to provide empowerment and support to inexperienced young mothers in the first two years after childbirth. Similarly, the MCYP program provides a meeting center where young parents below 25 years visit to seek information, advice, empowerment, and to socialize (Feijen-de Jong, 2022).
Lassi, Mansoor, Salam, Das, and Bhutta (2014) noted that to achieve the best pregnancy outcomes, appropriate interventions and support need to be developed and delivered during pre-pregnancy and throughout the pregnancy duration. To ensure that maternal care interventions are channeled on a case-by-case basis, there is the need to form segments that consist of the attributes and characteristics of pregnant women that frequently occur together. Pregnant mothers’ segmentation will ensure that counselling, preventive measures, and support strategies are designed and channeled according to the specific needs of each segment. Therefore, the objective of this study was to extract segments from existing data consisting of some prominent attributes and characteristics of pregnant women. The study employed a machine learning technique known as association rules mining to facilitate extraction of the segments.
Key Terms in this Chapter
Association Rule: A notation that shows the frequently occurring patterns among a set of items. The left-hand side of a rule shows the antecedent while the right-hand side shows the consequent.
Segmentation: An activity that subgroups individuals or objects in a particular domain into segments (clusters) that share a similarity or pattern.
Strength of a Rule: A rule is considered strong if it meets the threshold of support, confidence, and lift set by the modeler.
Maternal Risk Factors: These are the characteristics, features, and variables of a pregnant woman that pose a threat to the unborn child and the mother. Examples include teenage pregnancy, non-attendance to ante-natal care visits, smoking during pregnancy, and so on.
Maternal Attributes: These are the characteristics, features, and variables associated with a pregnant woman. Examples include age, weight, smoking status, attendance to ante-natal care visits, and so on.
Pregnant Women’s Segmentation: An activity that employs machine learning/data mining techniques to subgroup maternal attributes based on frequently occurring patterns.
Association Rules Mining: The process of employing machine learning/data mining techniques to extract the frequently occurring patterns from a given dataset.