This analysis examines the postpartum initiation of contraceptive use in Zimbabwe and its relationship with breastfeeding, postpartum amenorrhoea, and postpartum abstinence. The analysis utilizes longitudinal data collected in the monthly calendar of events of the 1994 Zimbabwe Demographic and Health Survey (ZDHS). ContraceptiveuseiswidespreadfollowingrecentbirthsinZimbabwe:nearly80percentofmothers startusingacontraceptivemethodwithintwoyearsfollowingabirth. Contraceptiveinitiationisconcentrated in the early postpartum period with 65 percent of mothers initiating contraceptive use within six months postpartum. At the same time, breastfeeding is almost universal and prolonged. The median duration of breastfeedingis18.7monthsand88percentofmothersarestillbreastfeedingat12monthspostpartum. The median duration o f postpartum amenorrhoea is 11.5 months, consistent with the long duration o f breastfeeding. In contrast, the median duration of postpartum abstinence is much shorter at only 3.9 months.
These findings suggest a widespread and prolonged overlap between contraceptive use and breastfeedingandpostpartumamenorrhoeainZimbabwe. Morethan70percentofrecentbirthsarefollowed by an overlap between contraceptive use and breastfeeding with 54 percent followed by an overlap between pilluseandbreastfeeding. However,mostpostpartumpilluseappearstobeofprogestin-onlypillsconsistent with current medical guidelines and the Zimbabwe Government health policy. Nearly 60 percent of births are followed by an overlap between contraceptive use and postpartum amenorrhoea and the median duration ofanoverlapwhenitoccursis12months. Thisfindingsuggestsconsiderableredundantuseofcontraception in Zimbabwe and raises questions about the efficiency of current postpartum family planning policy in light of concerns that redundant use of contraception may result in shorter birth intervals than might have been achievedunderalternativepolicies. However,ouranalysisshowsthatcontraceptivecontinuationratesare higher among women who initiate use early in the postpartum period than among women who initiate use later. This finding suggests that women who initiate use early in the postpartum period are likely to achieve similar birth intervals as women who initiate use later, although birth interval lengths were not examined'directly in this analysis. Overlap between contraceptive use and postpartum abstinence is less common and the median duration of an overlap when it occurs is three months.
Contrary to findings in other countries, we found no evidence of a relationship between postpartum contraceptive use and the duration of breastfeeding or postpartum amenorrhoea, but we found evidence of a negativeassociationbetweenpostpartumcontraceptiveuseandthedurationofpostpartumabstinence. The timingofcontraceptiveinitiationdoesnotcoincidewithweaning,exceptforaminorityofwomen. About10 percent of women start using contraception in the month they resume menstruation and around 21 percent initiate contraceptive use in the month they resume sexual relations.
Demographic and socioeconomic differentials exist in postpartum contraceptive behaviour and in breastfeeding, postpartum amenorrhoea, and postpartum abstinence patterns. Postpartum contraceptive use is more widespread and occurs earlier in urban areas compared with rural areas and among educated women. Older, high parity women are less likely to use contraception postpartum and fertility preferences appear to have little effect on postpartum contraceptive behaviour. Early postpartum contraceptive use is particularly common in the Harare/Chitungwiza province where 81 percent of mothers initiate contraceptive use within three months postpartum. This in turn results in very high levels of an overlap between contraceptive use and breastfeeding, postpartum amenorrhoea, and postpartum abstinence.