Clinical outcomes of postpartum intrauterine devices inserted by midwives in Tanzania

To assess the rate of complications following immediate postpartum insertion of intrauterine devices (IUDs) by trained midwives in Tanzania.


| INTRODUCTION
The intrauterine device (IUD) is an effective, reversible, long-term method of contraception. It is the preferred method used by 14.3% of modern contraception users globally, but by less than 2% of women of reproductive age in Sub-Saharan Africa. 1,2 With an estimated 160 million users worldwide, the IUD has been ranked the most widely used modern method of contraception. 3 Although the use of modern contraceptives (measured as contraceptive prevalence rate) has steadily increased over the past decade in Tanzania-from 20% in 2004-2005 to 27% in 2010 and 32% in 2015-2016-unmet need for family planning has remained unchanged at 22%-24% since 1999. 4 Currently, IUD use among married women remains low, estimated at 0.9% of married women. 5,6 Timing PPIUD insertion immediately after childbirth is considered ideal especially in low-resource countries where women do not return | 39 Muganyizi ET aL. for postnatal follow-up visits because of cost or distance. Postpartum IUD (PPIUD) practice has the potential to increase IUD uptake because immediately after childbirth there is no fear of an ongoing pregnancy, there is less risk of pain, no risk of interference with breastfeeding, it is easy to insert, and the woman and IUD provider are both available in the same setting, which reduces the time and cost of seeking interval IUD services. 7,8 Nevertheless, PPIUD insertion has been linked with increased risks of poor outcomes such as IUD expulsion, uterine perforation, and infection, when compared with interval IUD insertions. 9,10 Increased risks for poor PPIUD outcomes can be improved through standardized training and provider experience; this is supported by the Canadian Contraceptive Consensus, which is based on an extensive literature review. 1 In Tanzania, midwives do provide interval IUDs to clients but intrauterine manipulations in relation to childbirth are generally performed by clinicians. Moreover, preservice curricula for midwives do not provide sufficient knowledge and skills for them to perform PPIUD insertion. 11 In December 2015, the International Federation of Gynecology and Obstetrics (FIGO) introduced a program in Tanzania to institutionalize PPIUD through training of healthcare workers on counselling, provision of immediate PPIUD, advocacy to local health management teams, and information delivery to women during prenatal clinic and delivery services. 12 Task sharing of PPIUD insertion to midwives is relatively new in Tanzania although it is well known that delegation of some tasks to less-specialized healthcare workers improves access to services and interventions in resource-limited settings. 13 Since midwives are more readily available in delivery rooms than clinicians, they have a better opportunity to provide PPIUD and increase access to PPIUD services for women after delivery. The aim of the present study was to evaluate the safety of immediate PPIUD insertion measured by complications experienced within 6 weeks of insertion, by women whose providers were midwives. An additional aim was to discuss the implication of this task sharing in the light of complication rates in comparable studies in the literature when providers were clinicians.   Table 1).

| RESULTS
In total, 43 women reported a PPIUD-related complication by the end of the sixth week after IUD insertion, giving an overall complication rate of 7.2% (Table 2)

| DISCUSSION
The present study was conducted in the context of a crisis for human resources in health in Tanzania; across regions the human resource density ranged between 4 and 10 per 10 000 population, which is well below the global critical line of human resources for health of 23 per 10 000. 14 Given this context, task sharing of PPIUD insertions to T A B L E 1 Characteristics of women whose PPIUD insertions were conducted by midwives. T A B L E 2 Rates of PPIUD-related complications 6 weeks after insertion.

PPIUD complication No. (%)
Uterine infection in Turkey, the expulsion rate was 9.3% at the sixth week follow-up and the cumulative IUD removal rate was 10% compared with a 4.4% IUD removal rate in the present study. In contrast, whereas the providers in the present study were midwives who had just recently been trained on IUD insertion, in the Turkish study all insertions were done by experienced physicians using an IUD inserter.
The importance of experience was echoed by the consistently higher rates of complications among insertions done in Phase 2 compared with Phase 1 hospitals where the providers had relatively longer experience of practice-a trend that has also been noted by others. 1 Two studies in USA estimated higher expulsion rates of copper IUD (CuT380A) including Goldthwaite et al., 16 who reported a 20% rate at 12 weeks with 86% of these occurring within 6 weeks, and 17% at 4-8 weeks among trained physicians in Atlanta. 17 Other studies and reviews have reported higher IUD expulsion rates ranging from 9.5%-38% when an IUD is inserted within 10 minutes of delivery of the placenta, although the reference duration of follow-up was somewhat longer (3-12 months) than the 6 weeks in the current study. 10,[18][19][20][21] Although studies on postpartum complications following CuT380A IUD insertion are generally sparse, these cited comparable studies clearly indicate the relative safety of PPIUD insertion by midwives in Tanzania. In all of the cited studies, providers were physicians who had longer experience with PPIUD insertions or who had received training; therefore, the results for midwives in the present study, who had no prior experience, suggest a successful task sharing of PPIUD insertion services in Tanzania.
Spontaneous expulsion of IUD is usually considered to be due to puerperal uterine remodeling, although low insertion of the IUD could be a contributory factor. 15 15,20,24,25 this reaffirms that the complication rates in the literature are higher than the combined rates for the program in Tanzania.
This is one of few large studies that have prospectively followed-up women who have had CuT380A IUD insertion after vaginal delivery.
Most studies reporting outcomes of PPIUD complications have either been confined to levonorgestrel IUDs or other forms of copper IUD inserted intraoperatively at cesarean delivery. 25 There is little literature on the impact of midwives on PPIUD insertion. The strengths of the present study are its large sample size and focus on midwives who provide services to most delivering mothers in low-resource countries and the CuT380A IUD, which is available and affordable.
The study was limited because only half of the women with PPIUD insertions voluntarily came for follow-up at the program-affiliated clinics. Since it would have been unethical to force women who have benefited from the program to come for follow-up in the specified clinics, the study operated on the natural situation whereby women seek medical and reproductive healthcare services at the nearest location and where they believe they can get the desired service. To mitigate this bias, the selected follow-up clinics were geographically well distributed to capture women from diverse geographical locations.
Nevertheless, it is acknowledged that the selection of the study sample could have led to bias toward women with higher complication rates than the reality for the PPIUD program as the assumption is that women would be more likely to attend follow-up if they had a problem.
Another important limitation is a relatively short follow-up of 6 weeks, which limits comparability with other similar studies. Nevertheless, it is well accepted that most PPIUD complications occur within the initial 6 weeks after insertion. 16

CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.