Return to search

Understanding pregnancy intention in the Southern African setting: validation of the London measure of unplanned pregnancy in the Cape Town area, South Africa

Background: Developed in the United Kingdom, the London Measure of Unplanned Pregnancy (LMUP) is a psychometrically valid measure of pregnancy intention developed to capture the multifaceted and complex construct of pregnancy intention. A growing body of evidence suggests that unintended pregnancies are directly associated with poorer maternal and infant outcomes, as well as contributing to poorer longer-tern social, economic, educational and health outcomes for women, children, families and societies. An improved understanding of women's pregnancy intentions in South Africa is essential to reduce adverse consequences associated with unplanned and unwanted pregnancies, and to better prevention on unintended pregnancies. Such information could be used for the development of family planning, pre-conception and pregnancy planning, and ante/post natal care interventions. It could also be used to monitor the effects of interventions aimed at preventing unintended pregnancy. To this end, this research aimed to validate the LMUP for use in both the Afrikaans and Xhosa languages in South Africa. Methods: Three Xhosa and three Afrikaans speakers translated the LMUP, in which one translation was agreed upon for each translation. This translation was then back-translated and pre-tested on five to ten pregnant women using cognitive interviews. The measure was field tested with pregnant women who were recruited at two antenatal clinics and re-tested between one and two weeks after the initial interview. The data were analyzed using classical test theory, principal component analysis and hypothesis testing for both Xhosa and Afrikaans separately. Results: 150 women aged 18-42 (median 26.5), with parities of 0-3 (median 2) completed the Xhosa LMUP and 148 women aged 18-42 (median 28), with parities of 0-7 (median 1) completed the Afrikaans LMUP. There were no missing data for the Xhosa LMUP and less than 0.01% missing data for the Afrikaans LMUP. Scores ranging from 1-12, nearly the entire LMUP range, were captured in both Afrikaans and Xhosa .126 of 150 (84%) of the Xhosa and 105 of 148 (71%) of the Afrikaans were followed up for re-test, well in excess of the 50% target. The scale was internally consistent (Chronbach's alpha: Xhosa= 0.83; Afrikaans=0.72) and the test-retest data showed good stability (weighted Kappa: Xhosa = 0.82; Afrikaans=0.76) for both the Xhosa and Afrikaans LMUP. For the Xhosa LMUP, hypothesis testing confirmed that unmarried women (p = 0.0001) and women who were below 20 or 40+ (p = 0.008) were more likely to report unintended pregnancies. While median LMUP score was lower among women with 2-3 children (median LMUP score=4.5) compared to 0-1 children (median LMUP score=6), this difference was not statistically significant in this sample (p = 0.16). For the Afrikaans LMUP, hypothesis testing confirmed that unmarried women (p = 0.0001), women who were <25 or 40+ (p=0.005) and women with fewer children (p=.003) were more likely to report unintended pregnancies. The LMUP in both languages was re-analysed without the question on contraception use. This reduced the range of LMUP scores from 1-12 to 0-10 for both Xhosa and Afrikaans with median scores of 4 and 6 for Xhosa and Afrikaans, respectively. Cronbach's alpha increased to 0.87 for Xhosa and 0.74 for Afrikaans and all items loaded on to one component with Eigenvalues of 3.3 and 2.6, for Xhosa and Afrikaans respectively. Conclusion: The Xhosa and Afrikaans-language LMUPs are valid and reliable measures of pregnancy intention, which can now be used in South Africa in research, surveillance/monitoring and clinical care.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/23705
Date January 2016
CreatorsErnstoff, Elizabeth
ContributorsConstant, Deborah
PublisherUniversity of Cape Town, Faculty of Health Sciences, Women's Health Research Unit
Source SetsSouth African National ETD Portal
LanguageEnglish
Detected LanguageEnglish
TypeMaster Thesis, Masters, MPH
Formatapplication/pdf

Page generated in 0.0039 seconds