Uganda’s total fertility rate (TFR) at 6.2 is one of the highest in sub-Saharan Africa and globally[1]. Uganda more so, has a high maternal mortality ratio (MMR) at 435 maternal deaths per 100,000 live births and an infant mortality rate (IMR) of 54 per 1,000 live births per year[2]. High maternal morbidity and mortality could partly be attributed to unintended pregnancies, short birth intervals and higher risk of obstetric and newborn complications associated with low contraceptive use[3]. About 44% of pregnancies in Uganda are unintended[2] with occurrence of unsafe abortions estimated at 62 per 1,000 women aged 15–49 years[4]. These undesirable maternal and child health outcomes associated with high TFR could be substantially reduced by meeting the family planning (FP) needs of women in developing countries[5]. Provision of highly efficacious family planning (FP) services contributes to a reduction in maternal mortality by lowering the risk of maternal death per birth hence preventing high-risk and high-parity births[6]. It also offers individuals and couples’ ability to anticipate and attain the desired number of children by birth spacing and timing[7].

Use of long acting reversible methods is proposed as a strategy to reverse undesirable maternal health consequences in developing countries[8, 9]. Scientific evidence has determined implant and copper-bearing Intra-Uterine Device (IUD) contraceptives to be highly effective and well tolerated[10, 11]. Similarly, injectables and implants are proven to be safe, effective and reversible contraceptive options[12, 13]. Despite evidence of LARC effectiveness and safety, actual uptake in resource-poor settings like Uganda is low, and is possibly affected by several factors such as knowledge and general awareness of contraceptive methods[14], access to different contraceptive methods, user characteristics, technology[15] and socio-economic status[16].

Currently in Uganda, the contraceptive prevalence rate (CPR) is 30% and LARC prevalence is as low as 13% among all women[2, 17]. The prevalence of IUD use is lowest at 0.4%, implants as low as 1.9% and injectable contraceptives at 10.7%[2]. Kampala district, the study site, has a LARC prevalence of 22.7% with implant (1.6%), IUD (1.8%) and injectables (19.3%)[2]. This study set out to determine the reasons for low use of LARC in Lubaga division, Kampala, particularly assess knowledge and attitudes among women of reproductive age use of LARC attending reproductive health services.

A health facility-based, cross-sectional study was conducted for a period of nine weeks between March and April 2012 in Lubaga division. Lubaga is one of five administrative divisions in Kampala district, the capital of Uganda. The total population of the area is 411,900 with 53.6% females[18] and about 92% lives in peri-urban locations[19]. Lubaga division’s health structure is currently comprised of 124 registered health facilities and these are in three categories; 2 Public facilities, 120 private for profit and 2 Private Not-For-Profit (PNFP) which all provided reproductive health services including LARC services. The participants were both FP users and non-users who were females 15–49 years attending any reproductive health service (i.e. family planning, post-natal and outpatient clinics) in Lubaga division. A family planning user was defined as a woman who at the time of the interview was currently using any contraceptive method[2]. A non-FP user was defined as a woman who at the time of the interview had never used any contraceptive or had previously used a contraceptive but discontinued its use.

Data was entered, cleaned using Epi info 3.5.1 and statistical analysis done with STATA Version12.0. Data was collected using semi-structured interviewer administered questionnaires which contained information on independent variables. During administration of interviews, the respondents were not prompted with any contraceptive pictorials or FP product names. The outcome variable was binary i.e. current user or non-user of LARC methods. Prior to data collection, questionnaires were pre-tested at a non-participating health centre III in Nakawa division in Kampala that was purposively selected due to similarity of participants’ socio-cultural characteristics in order to determine their suitability for collecting study data and then editing of the data collection tools appropriately. Field editing of questionnaires was conducted by the supervisors on daily basis to check for completeness.

The proportion of participants currently using LARC methods was 31.7%. Among 565 study participants, about 21% (120 out of 565) reported having ever had an abortion. About 57% (325 of 565) reported having used at least one LARC method prior to current use of LARC at the time of the interview.

Knowledge of administration site for IUD, implant and injectables was 75.9%, 80.2% and 91.9% respectively. Knowledge levels of administration site of specific LARC methods were high among current use of LARC i.e. 82.7%, 88.3% and 95.5% for IUD, implant and injectables respectively (Table 2).

Approximately half (48.1%), 272 of 565 respondents agreed that the contraceptive to use is their partners’ decision.

On the other hand, current use of LARC methods was significantly reduced if the respondents agreed that LARC should be used by married women (PRR 0.62; 95% CI 0.43, 0.92; p = 0.016).

The aim of this study was to assess the level of knowledge and attitudes towards LARC methods as well as factors associated with its use among women of reproductive age attending reproductive health services in Lubaga, division Kampala. Whereas the level of knowledge of specific LARC methods in Lubaga division (Table 2) was slightly higher than that of Kampala district[2], it is still sub-optimal. However across the three LARC methods, women had higher levels of knowledge about injectable contraceptives than implants and IUDs. Knowledge of duration of protection from pregnancy was higher for injectable contraceptives than implant and IUD. Knowledge of site of administration for all LARC methods was also relatively high comparable to the national knowledge levels for the three LARC methods[2] suggesting that the needed FP information is reaching the women in Lubaga. The high level of knowledge of LARC methods in our study may be explained by the highly selective group of females selected from health facilities. Lubaga, a peri-urban setting also has a high number of health units which is likely to increase physical access to FP services including contraceptive counseling. However, there is need for further research into why the high level of knowledge does not translate into actual higher use of LARC in this urban setting[2].

Low use of LARC may be due to misconceptions about causation of permanent infertility (Table 3). A survey in Ghana, revealed that women had high knowledge levels of IUD but cited its side effects as the main reason for non-use[27]. Similarly, studies in Kenya and Nigeria revealed that among injectable and implanon (implant) users respectively, misconceptions[28], side effects[28, 29] and husbands’ opposition[28] were main reasons for their discontinuation.

In contrast, a study in Jimma, Ethiopia, revealed that women were least knowledgeable about IUD and more knowledgeable about implants and injectables especially among married couples[30]. The difference in findings could be explained by the fact that the Ethiopian study was conducted among married couples which could have been a highly self-selected group that had jointly agreed to use LARC.

In a study in rural Ethiopia, knowledge of IUD was much lower at 13.1%[31] when compared with knowledge of IUD in this study in Lubaga division. Though knowledge of injectable contraceptives (97.8%) is nearly universal when compared to this study in Lubaga, knowledge of implants (74.4%) was comparable[31]. These study differences in knowledge of IUD and implants could potentially be due to rural–urban differences in socio-demographics and availability of specific LARC methods. Since injectable contraceptives are the predominant LARC method currently used in both studies, this may suggest that injectable contraceptives are the LARC method mostly available. Provision of injectable contraceptives at both community and facility levels when compared to facility provision of implants and IUD in some sub-Saharan countries like Nigeria[32], Madagascar[33], Ethiopia[34] and also Uganda[35, 36] could have contributed to high injectable contraceptive knowledge and current use.

Nearly half (48.1%) of the women in this study thought that their male partners should decide on the contraceptive to use (Table 3) suggesting that socio-cultural perceptions play an important part in the contraceptive choices made. A study in Kenya and Ethiopia found that partner approval[37] and husband support[38] influences use of FP services and contraceptives respectively. Our study in Lubaga had similar findings on partner’s decision on spouses’ choice of contraceptive. Partner decision was associated with women’s contraceptive use similar to findings from Ghana, Nigeria and Rwanda where partner approval, partner’s support[39, 40] or partner objections[41–43] towards contraceptive choice influenced contraceptive use. In South Africa, women that strongly agreed to male decision making regarding child-bearing did not influence effective contraception[44]. In Uganda[45, 46] and Burkina Faso[47], scientific evidence has also shown that positive influence of male partners may affect maternal and child health outcomes. Nearly half of the women in this study thought that their male partners should decide the contraceptive to use therefore socio-cultural perceptions in Lubaga especially related to roles of males and females in decision making in families might explain these attitudes among women.

In Nigeria,[48] and Bangladesh[42], acceptors of long acting contraception were more likely to be married women. A study in Nigeria consistent with our study showed that almost half of the married women had previously used IUD and injectables[49]. Since comparable proportions of use of LARC in all three studies were married women, this may suggest that they may have good attitudes towards using LARC methods.

The findings of our study should be interpreted in light of some limitations; the study was cross-sectional therefore we could not establish temporality between current use of LARC and the independent factors. Since the study relied on the respondents’ self-report, there could have been potential for recall bias about the history related to use of LARC methods. We attempted to control for potential confounders of known factors in the multivariable analysis. Misclassification bias may have been introduced when collapsing attitude items from a 5 to 3 point likert scales, the new categorization assumed that those reporting “unsure” disagreed. However, given the small number of responses in the “unsure” category this is unlikely to have greatly biased the results. Suggestion for future research could include conducting qualitative data collection methods in order to adequately assess behaviors of women associated with contraceptive use.

This study revealed a relatively low level of current use of LARC among women attending clinics in Lubaga, Kampala. Knowledge of site of administration of implants, previous use, and women’s attitude about the important role of male partners in their choice of contraceptives was associated with current use of LARC. Strategies to strengthen client education may be integrated within reproductive health programs in Lubaga division in order to dispel possible myths about LARC. Involving males in the decision making process of contraceptives may also be integrated into strategies to promote family planning services.