The vast majority of the postabortion women accepted postabortion contraceptive services and stated they were using contraception 12 months after the abortion. Follow-up information, however, was only achieved among 59 percent of the included women and the long-term contraceptive adherence among the 41 percent of the women who were lost to follow-up is unknown. The fact that abortion is illegal in the studied setting is likely to have contributed to the high drop-out rate. Further, whether illegal or not, an induced abortion is often considered a traumatic or even stigmatising event, which women would like to avoid being confronted with at a later stage in their life – a fact that is likely also to have had a negative impact on the follow-up rate. Similar problems with high drop-out rates have been experienced in other studies focusing on the long-term impact of postabortion care .
The high drop-out rate experienced in the present study raises the question of the extent to which the women who contributed with follow-up data differ systematically from those who did not. Loss to follow-up was more pronounced among young, unmarried women who had not given birth previously. In Tanzania, as well as in many other low income societies, premarital sex is considered immoral , and for fear of their abortion experience being disclosed, young or single women may thus have been more likely to provide wrong addresses to avoid a home visit, even though they accepted such a visit when they were enrolled in the study. In addition to the abortion related stigma, a high migration rate is likely also to have had a negative impact on the proportion of women who contributed with follow-up data. In Temeke Municipal, many settle in unplanned squatter areas with social problems and are likely to move when a better housing possibility becomes available, a situation which is reflected in a high yearly immigration rate of 25 percent . This situation that is likely to be more pronounced among young and single women, who have not yet established themselves with their own family, may also partly explain the low follow-up rate achieved.
The findings form the present study, although hampered by a high drop-out rate, adds to the evidence that women having experienced unsafe abortions are likely to accept and use contraception. One explanation behind the high contraceptive prevalence rate found in this study may be that the service was offered as a ward based activity by well trained counsellors. An assumption which is supported by a Kenyan study where postabortion contraceptive service was accepted by 75 percent when it was on the ward, while only 41 percent obtained a method when asked to visit a separate site within the same hospital after discharge . In the present study, the women were further asked to return to the hospital every 3 month to discuss their contraceptive use and have experienced problems addressed. This approach may additionally have had a positive impact on the women's contraceptive adherence. Hence, women who have experienced an induced abortion may be reluctant to use contraception due to lack of detailed knowledge about contraception and fear of experienced or imagined side effects [20, 21]. Frequent counselling that addresses the women's concerns may therefore increase the women's contraceptive adherence. Whether the approach of inviting the women to return to the hospital for ongoing counselling and service is the most appropriate way to strengthen the women's ongoing contraceptive use can be discussed. Hence, at a programmatic level, the cost-effectiveness of such an approach may be questioned and it may be more relevant to refer the women to a primary health facility.
Postabortion contraceptive services are apparently well accepted by women who are admitted with complications after an unsafe abortion and should thus be considered an important means in addressing the problem of unsafe abortions. However, while hospitals incur only minor additional costs by introducing postabortion contraception, few have done so successfully. To begin with, contraceptive services and incomplete abortion treatment are typically delivered and administered separately – in different areas of one building, in separate buildings, or even within different institutions. In addition to separately administered services, lack of appropriate training, time, and institutional support can make physicians and nurses in the wards where patients receive postabortion care reluctant or unable to take on the duty of providing contraceptive services. To ensure postabortion contraceptive services are being offered in a comprehensive and sustainable way the following requirements should be addressed: qualified contraceptive counsellors should be present in the gynaecological ward; the most commonly used contraceptive methods (oral contraceptives, injections and condoms) should be made available in the ward and; the service should be monitored and evaluated through a simple routine supervision system.
High HIV prevalence rates, ranging from 19 percent to 25 percent, have been found among Tanzanian single women who have experienced an unsafe abortion . These figures are considerable higher than sentinel surveillance results reporting HIV prevalence rates of 10–14 percent among pregnant women in Temeke Municipal. Therefore, in Tanzania as well as in other areas with high HIV prevalence rates, counselling about HIV and condom use should be considered an essential aspect of postabortion care. The present study documents that such counselling may lead to an increased number of single women practising safe sex. It has been documented that the vast majority of Tanzanian women are aware of risky behaviour and the protective role of abstinence, faithfulness to one uninfected partner and condom use . However, as the present study shows, they face a number of obstacles in translating their knowledge into safer sex practices. Both women and men associate condoms with promiscuity, STIs and HIV. Suggesting condom use may, therefore, imply either that one has a sexually transmitted disease or that one mistrusts one's partner . This association of condom use with promiscuity undermines the use of condoms as an HIV-preventive measure, especially among the married women. These difficulties related to condom use should be acknowledged when counselling married women on contraceptive use. Especially, since marriage has been found to be a significant risk factor for HIV infection in Tanzania as well as in other sub-Saharan African countries .