misoprostol distribution during antenatal care visits ... · pdf filemisoprostol distribution...

December 2009

Misoprostol Distribution during Antenatal Care Visits

Preliminary Report

Misoprostol Distribution during Antenatal Care Visits: Preliminary Report


Table of Contents

List of Tables and Figures ............................................................................................................................ ii Acronyms .................................................................................................................................................... ii 1. Introduction ............................................................................................................................................ 1

1.1 Background ....................................................................................................................................... 1 1.1.1 Postpartum Hemorrhage in Tanzania ........................................................................................ 1 1.1.2 Misoprostol for Prevention of PPH ............................................................................................ 1

1.2 Rationale for Misoprostol Distribution at ANC for Prevention of Postpartum Hemorrhage at Home Births ............................................................................................................................................ 2 1.3 Partnerships ...................................................................................................................................... 2 1.4 Goals and Objectives ......................................................................................................................... 2

2. Methods .................................................................................................................................................. 3 2.1 Strategy and Design .......................................................................................................................... 3 2.2 Project Personnel and Training ......................................................................................................... 4

2.2.1 Organizational Structure ............................................................................................................ 4 2.2.2 Training Structure ...................................................................................................................... 4

2.3 Location and Participants .................................................................................................................. 4 2.4 Data Collection and Management .................................................................................................... 5 2.5 Ethical Review ................................................................................................................................... 5

3. Results..................................................................................................................................................... 5 3.1 Development of Information, Education, and Communication Materials ........................................ 5 3.2 Enrollment ........................................................................................................................................ 6 3.3 Participants’ Antenatal and Delivery Characteristics ........................................................................ 7 3.4 Feasibility: Misoprostol and IEC Message Coverage ......................................................................... 8

3.4.1 Coverage of IEC Messages .......................................................................................................... 8 3.4.2 Coverage of Misoprostol Distribution ...................................................................................... 10

3.5 Program Effectiveness: Comprehension of IEC Messages and Coverage of Misoprostol at Home Births..................................................................................................................................................... 11

3.5.1 Comprehension of IEC Messages ............................................................................................. 11 3.5.2 Use of Misoprostol at Home Births and Protected Births ........................................................ 12 3.5.3 Bleeding-related referrals and the need for additional interventions...................................... 14

3.6 Safety: Correct Use of Misoprostol and Postpartum Symptoms ..................................................... 15 3.7 Acceptability: User Perspectives on Misoprostol Use ..................................................................... 16

4. Conclusions ........................................................................................................................................... 16 5. Recommendations ................................................................................................................................ 18 6. References ............................................................................................................................................ 19

i


List of Tables and Figures Figure 1: Organizational structure of the project (including one of four districts) ..................................... 4 Table 1: Population, estimated births, ANC coverage, and ANC resources in project districts ................... 4 Table 2: Enrollment, March – August 2009 ................................................................................................. 7 Table 3: ANC and Delivery characteristics (n=1,316) .................................................................................. 7 Figure 2: Location of delivery by month (n=1,316) ..................................................................................... 8 Table 4: Community outreach exposure (n=1,316) ..................................................................................... 9 Figure 3: Most important source of misoprostol information by district (n=1,316) ................................. 10 Table 5: Coverage of misoprostol distribution to enrolled clients at ANC visits (n=4,829) ....................... 10 Figure 4: ANC attendance, enrollment, ANC attendance at 32 weeks gestation, and misoprostol distribution (n= 4,829) .............................................................................................................................. 11 Figure 5: ANC attendance at 32 weeks or greater gestation..................................................................... 11 Table 6: Comprehension of IEC messages (n=1,316) ................................................................................ 12 Table 7: Coverage of misoprostol at home births ..................................................................................... 13 Table 8: Coverage of protected births (n=1,316) ...................................................................................... 13 Figure 6: Coverage of uterotonic for PPH prevention (n=1,316) ............................................................... 14 Figure 7: Coverage of protected births at home vs. health facility deliveries (n=1,316) ........................... 14 Figure 8: Bleeding-related referrals and need for additional interventions .............................................. 15 Table 10: Correct use of misoprostol at home births (n=463) .................................................................. 15 Table 11: Reported experience of postpartum symptoms (n=1,316) ....................................................... 16 Table 12: Acceptability of misoprostol ...................................................................................................... 16

Acronyms ANC Antenatal Care

CORPS Community Resource Persons

IHI Ifakara Health Institute

IRB Institutional Review Board

MOHSW Ministry of Health and Social Welfare

PDA Personal Digital Assistant

PPH Postpartum Hemorrhage

TBA Traditional Birth Attendant

TBC Tanzania Broadcasting Corporation

UCB University of California, Berkeley

VSI Venture Strategies Innovations

ii


1. Introduction

1.1 Background

1.1.1 Postpartum Hemorrhage in Tanzania An estimated 13,000 women die each year due to pregnancy- and delivery-related causes in Tanzania, translating to a high maternal mortality ratio – 578 to 950 deaths per 100,000 live births(NBS Tanzania and ORC Macro 2005; WHO 2005), but one that is comparable to Africa’s as a whole (900 deaths per 100,000 live births) (WHO 2005). Postpartum hemorrhage (PPH), or bleeding in excess of 500 mL after childbirth, is the leading cause of maternal death in Tanzania.

It is impossible to predict who will experience PPH. The average time to death from onset of PPH is two hours so any delay in seeking health care can be deadly (Maine 1993). Since many women do not present with risk factors, PPH prevention is extremely important, especially in Tanzania where women may have limited access to care and over half of deliveries take place outside of a health facility (NBS Tanzania and ORC Macro 2005).

There are several technologies that have proven effective at managing PPH. However, current methods of PPH prevention and treatment – uterotonic agents such as oxytocin and ergot preparations that cause the uterus to contract and reduce postpartum bleeding – must be administered by injection, and require refrigeration to preserve their potency. Consequently, there is a need for an effective method of preventing PPH, which could be an alternative drug for prevention of PPH when deliveries take place at home.

1.1.2 Misoprostol for Prevention of PPH Misoprostol is a safe, effective and low-cost tablet that has been shown to reduce postpartum bleeding after delivery. While injectable uterotonics are more effective than misoprostol, current evidence supports the use of misoprostol where oxytocin is infeasible (Joy, Sanchez-Ramos et al. 2003; Geller, Adams et al. 2006; Lagenbach 2006; Alfirevic, Blum et al. 2007). Furthermore, several features of misoprostol make it more feasible than oxytocin in areas where maternal mortality is the highest (Derman, Kodkany et al. 2006). Misoprostol is in tablet form, and thus has several routes of administration that do not require a skilled health provider. The tablet is stable in field conditions and has a long shelf life (Tang, Gemzell-Danielsson et al. 2007).

Therefore, misoprostol has the potential to greatly reduce the number of maternal deaths due to PPH in a country such as Tanzania. Indeed, the Tanzanian Federal Drug Regulatory Agency approved the use, sale and distribution of misoprostol for use in the prevention and treatment of PPH in September of 2007. Furthermore, several countries in Africa have already included misoprostol for the prevention of PPH in their lists of essential medicines. A number of international agencies are now working in partnership with ministries of health around the globe to introduce misoprostol for PPH prevention and to train providers in its safe administration.

1


1.2 Rationale for Misoprostol Distribution at ANC for Prevention of Postpartum Hemorrhage at Home Births Various studies have been conducted to prove the effectiveness of misoprostol in reducing cases of PPH, the need for additional interventions and the need for referrals to facilities (Prata, Mbaruku et al. 2005; Walraven, Blum et al. 2005; Derman, Kodkany et al. 2006; Prata, Gessessew et al. 2009). These studies were hospital-based, as well as community-based using assistant nurse-midwives, traditional birth attendants (TBAs), or community volunteers to dispense and/or administer misoprostol during deliveries that take place at home. However, in settings such as Tanzania, especially in the project districts (Kigoma, Kilombero, Rufiji and Ulanga) where 94% of women receive antenatal care from a health professional at least once during pregnancy yet over half (53%) of deliveries take place at home (NBS Tanzania and ORC Macro 2005), a measurable impact of the use of this drug would be possible if women could receive information on how to take misoprostol and be offered the drug during antenatal care visits. This approach would enable those who cannot reach a facility to deliver to have access to this life-saving technology.

1.3 Partnerships Ifakara Health Institute (IHI) is an autonomous, non-for-profit organization registered in Tanzania focusing on health research. The mission of IHI is to develop and sustain a district-based health research and resource centre capable of generating new knowledge and relevant information for public health policy and actions. IHI leads and oversees the implementation of this operations research project, including ongoing monitoring of activities and data management.

Venture Strategies Innovations (VSI) is a California-based nonprofit organization committed to improving women’s health in developing countries by creating access to effective and affordable technologies on a large scale. Our innovative approach involves partnerships that build upon existing infrastructure, resources, and markets. We focus on reducing barriers to access and enhancing human capacity to bring about sustainable improvements in health. VSI provided financial and technical support to this project, including the development of data collection tools, training materials, monitoring and evaluation design, and data analysis management.

Bixby Center for Population, Health, and Sustainability is a research center located at the University of California, Berkeley (UCB) School of Public Health. The Center is dedicated to developing innovations to improve reproductive health in resource-poor settings, including reliable health information systems, local access to essential technologies, and guidelines for prioritizing interventions to maximize health impact. The Center assists in the implementation of maternal health programs and seeks to improve the health outcomes of the world’s poorest and most vulnerable women and their families. With VSI, the Center provides technical assistance to this project.

1.4 Goals and Objectives The goal of this project is to save mothers’ lives by preventing postpartum hemorrhage (PPH) at home births with misoprostol tablets among women who are unable to reach a facility to deliver. To reduce the number of women who die due to excessive bleeding at home births, this project

2


distributes misoprostol tablets at antenatal care (ANC) visits to women who cannot reach a facility to deliver. Distribution of misoprostol is accompanied by a community awareness campaign on birth preparedness and PPH prevention.

The project’s main objective is to demonstrate safety, acceptability, feasibility, and program-effectiveness of misoprostol distribution through ANC care visits. More specifically, this project aims to:

• Provide necessary evidence that women can safely self-administer misoprostol for prevention of PPH at home births after being educated on and receiving the drug at ANC visits.

• Determine whether women and any potential labor support members of their family or community find misoprostol to be an acceptable means of preventing PPH at home births.

• Demonstrate that the ANC visit is a feasible and effective means of distributing misoprostol for PPH prevention to women who cannot get to a facility to deliver and consequently, give birth at home.

Evidence generated from this project will inform Tanzania’s policy decision makers on the use of misoprostol in home births, dispensed during antenatal care visits. Results from this project will demonstrate that this model can be applied throughout Tanzania, especially in settings where most of the women deliver without a skilled assistant.

2. Methods

2.1 Strategy and Design This operations research is comprised of two components:

1. Community awareness campaign on birth preparedness and PPH prevention: Using various methods of communication, including radio spots and in-clinic posters and pamphlets in Kiswahili and English, the campaign creates awareness about the risks of PPH, the importance of delivering in a health facility if possible, blood loss measurement with the kanga, and the use of misoprostol for the prevention of PPH. The awareness campaign focuses on the importance of ANC visits throughout pregnancy and that after 32 weeks women are eligible to receive misoprostol at ANC visits. The campaign utilizes radio messaging, print materials, and person-to-person interaction to communicate the main messages of the campaign, with primary emphasis on the project’s tagline, “Plan early for a safe delivery.”

2. Focused ANC with misoprostol distribution: Education provided at ANC visits is a cornerstone of this project and ANC serves as the entry point for pregnant women to enroll in the project and receive misoprostol tablets. During routine ANC care, providers conduct an Education Session on safe delivery and PPH prevention, including information on misoprostol, to all women. During the following one-on-one sessions, providers ask women if they would like to participate in the project, enroll assenting women by asking them to sign the informed consent form, provide further information on the use of misoprostol for the prevention of PPH, and dispense misoprostol tablets to women with 32 weeks or more gestation for use at home births in the event that they cannot deliver in a health facility. Women can enroll in

3


the project at any point during their pregnancy, but only receive misoprostol when they have reached 32 weeks gestation. Enrolled women are asked to participate in a postpartum interview, either at the health facility or at home, to learn about their delivery experience and knowledge and use of misoprostol.

2.2 Project Personnel and Training

2.2.1 Organizational Structure The lead investigators of this project are Dr. Godfrey Mbaruku, Director of the Empower Project at the IHI, and Dr. Ndola Prata, Assistant Adjunct Professor at UCB School of Public Health, Scientific Director of the UCB Bixby Center, and Medical Director of VSI.

Staff at both the IHI and VSI assisted in the development, coordination, and organization of project training, implementation, and data management. In addition, each district has a District Coordinator who oversees ANC providers, research assistants, and community resource persons (CORPs) and TBA supervisors, who in turn manage the CORPs and TBAs working in each district (Figure 1).

2.2.2 Training Structure In November 2008, IHI and VSI led a five-day master training on project protocols for project management, District Coordinators, and Research Assistants. Following the master training, each district held trainings for ANC providers led by the District Coordinator and project management. In addition, each district held trainings for TBA and CORPs supervisors, and TBA and CORPs on the community awareness campaign. All project staff were trained by mid-January 2009, including 282 ANC providers, 40 Research Assistants, 165 CORPS, 23 CORP supervisors, 150 TBAs and 18 TBA supervisors.

2.3 Location and Participants The project is being conducted in four districts in Tanzania totaling around one million in population: Kigoma Urban, Kilombero, Ulanga and Rufiji. All facilities in these four districts that provide ANC participate in the project. Table 1 specifies the population, ANC coverage, and resources in each of the project areas.

Table 1: Population, estimated births, ANC coverage, and ANC resources in project districts

Sites Population Estimated

Births/Year ANC Coverage # ANC sites # ANC providers

Figure 1: Organizational structure of the project (including one of four districts)

4


Kigoma Urban 168,112 6,725 91% 13 36 Kilombero 399,600 3,675 98% 33 103 Ulanga 212,597 5,527 87% 33 32 Rufiji 226,928 4,312 98% 47 111 Total 1,007,237 20,239 -- 126 282

2.4 Data Collection and Management Data is collected on every woman attending ANC during her first visit and at subsequent visits for participants who are enrolled in the project using the Misoprostol Addendum. This form records basic ANC information, enrollment, and misoprostol distribution to the women who are 32 weeks gestation or greater.

During enrollment, ANC providers will ask participants in the project to participate in a follow-up postpartum interview. The Postpartum Interview Questionnaire collects participants’ views of and experience with misoprostol, knowledge about PPH and misoprostol, and delivery experience, which will provide key data on project outcomes and are an essential component of the project. The postpartum interview is conducted either before discharge if a participant delivers at a health facility, when the participant returns to the health facility for postnatal care, or through active follow-up.

Data entry of project data collection forms is ongoing, conducted by Research Assistants using personal digital assistants (PDAs). Research Assistants meet with their respective District Coordinator on a regular basis to upload their inputted data, and the District Coordinator transfers district databases to the Data Manager at IHI on a monthly basis. Review, management, and basic analysis of the data are conducted jointly by the IHI Data Manager and the VSI M&E Coordinator.

2.5 Ethical Review Institutional Review Board (IRB) approval for this project was obtained from the University of California, Berkeley and the National Institute for Medical Research and Ministry of Health and Social Welfare (MOHSW), Tanzania. Women are asked for their informed consent at enrollment and before conducting the postpartum interview.

3. Results

3.1 Development of Information, Education, and Communication Materials In conjunction with the education that women receive at ANC visits, an extensive information, education and community awareness campaign on birth preparedness and PPH prevention is ongoing in project areas to bolster safe delivery messages, the importance of delivering in a facility, and women’s knowledge of misoprostol for PPH in the project communities.

The community awareness campaign includes four key messages:

1. Importance of delivering at a health facility; 2. Birth preparedness and planning early for a safe delivery; 3. PPH causes and consequences, and recognition using the kanga method;

5


4. Misoprostol for the prevention of PPH, which is available at ANC clinics once pregnant women have reached 32 weeks gestation.

Two radio scripts of 30 seconds in length were recorded in Kiswahili and aired on Pambazuko and Tanzania Broadcasting Corporation (TBC) FM radio stations. Pambazuko radio is a local FM radio station that broadcasts from Ifakara town and covers Kilombero and Ulanga districts. For the first three months of the project, seven radio spots were aired per day. TBC has national coverage, and for the purposes of this campaign targeted Kigoma urban and Rufiji district, and also covered some areas of Ulanga district that were not covered by Pambazuko. During the first month, two radio spots aired per day on TBC, totaling 60 radio spots. In the second and third months, two radio spots aired every two days; thus, 22 radio spots were aired per month in the second and third months.

All women coming to ANC in the project districts are provided an Education Session on birth preparedness and PPH prevention that includes information on the four main messages of the awareness campaign. To increase community-level awareness on birth preparedness, PPH, and misoprostol, CORPs and TBAs conduct awareness meetings with community leaders and women’s groups in their geographic areas, and conduct one-on-one information sessions with pregnant women in their villages.

IHI, VSI and the MOHSW collaboratively created print materials for both clinic and community-level education. Print materials include informational posters to hang in the clinic and pamphlets for the client to take home with her, both of which display pictorial instructions on how to use misoprostol for prevention of PPH. 2,000 informational posters were printed and distributed evenly between the districts. In addition, 20,000 prints of the pamphlets were produced and distributed to the districts in proportion to their respective population size (Kigoma = 4,000; Rufiji = 6,000; Kilombero = 5,000; Ulanga = 5,000).

A promotional poster was also developed for community awareness, and the CORPs and TBAs use the pamphlets when conducting one-on-one educational sessions with pregnant women. 8000 copies of this poster were produced and distributed equally between the districts.

3.2 Enrollment Enrollment of women began during the first two weeks of February 2009, with all four districts reaching full implementation of the project by the second week of February. While implementation and data collection will continue until the end of December 2009, complete Misoprostol Addendum data is available through August from all four districts. Collection of postpartum interviews is ongoing, and data is currently available for approximately 27% of those enrolled between March 1 and August 31, 2009.

Of the women who attend ANC, we anticipated that 80% would agree to participate in the project resulting in approximately 7,200 enrollees in six months. ANC providers have been very successful in recruiting women to the project, achieving a 99% enrollment rate of women coming to ANC and recruiting 8,433 women to participate in the project.

The following analyses are based on six months of Misoprostol Addendum data (March to September 2009) and currently completed and entered Postpartum Interview Questionnaires of women who enrolled between March 1 and August 31, 2009, and who had an expected date of delivery before September 1, 2009. We therefore analyzed the Misoprostol Addendum information

6


of 4,829 participants, over a quarter of whom had completed a postpartum interview (n=1,316; 27%) (Table 2).

Table 2: Enrollment, March – August 2009 Kigoma

N (%) Kilombero

N (%) Rufiji N (%)

Ulanga N (%)

Total N (%)

ANC Attendance 1,248 2,501 2,795 1,967 8,511 Number of clients enrolled in project of those who attended ANC

1,237 (99.1) 2,493 (99.7) 2,785 (99.6) 1,918 (97.5) 8,433 (99.1)

Included in preliminary analysis1

845 (68.3) 1,380 (55.4) 1,688 (60.6) 916 (47.8) 4,829 (57.3)

Completed postpartum interview

125 (14.8) 225 (16.3) 707 (41.9) 259 (28.3) 1,316 (27.3)

1Enrolled between March 1 and August 31, 2009 with an expected date of delivery before September 1, 2009

3.3 Participants’ Antenatal and Delivery Characteristics Table 3 presents basic ANC and delivery characteristics of participants included in this preliminary analysis based on data collected from the postpartum interview. Participants attended an average of slightly fewer than three ANC visits over the course of their pregnancy (2.9).

Overall, almost two thirds of women delivered in a health facility (61%), which is slightly higher than the national average of 47% in the most recent Demographic and Health Survey(NBS Tanzania and ORC Macro 2005). There was some variation in the rate of health facility delivery between districts, with Rufiji reporting the lowest rate of facility births (55%) and Kilombero reporting the highest (72%).

Correspondingly, approximately two thirds of women delivered with a skilled provider, most often a nurse-midwife (51%). However, there is a missing response on attendant at delivery for 515 participants, 508 of whom delivered either at home or on the way to the health facility, where it is presumable that they delivered with unskilled providers (e.g. TBA, friend, relative, or alone).

Table 3: ANC and Delivery characteristics (n=1,316) Kigoma

N=125 Kilombero

N=225 Rufiji

N=707 Ulanga N=259

Total N=1,316

Average number of ANC visits (+SD)

3.06 (+ 0.07)

3.33 (+ 0.07)

2.92 (+ 0.03)

2.61 (+ 0.05)

2.94 (+.025)

Location of delivery P<0.001 Home 40 (32.0) 55 (24.4) 303 (42.9) 81 (31.3) 479 (36.4) Health facility 82 (65.6) 163 (72.4) 391 (55.3) 172 (66.4) 808 (61.4) On the way to health facility

3 (2.4) 7 (3.1) 13 (1.8) 6 (2.3) 29 (2.2)

Attendant at delivery P<0.001 Doctor 17 (13.6) 12 (5.3) 26 (3.7) 5 (1.9) 57 (4.3) Assistant Medical 14 (11.2) 12 (5.3) 26 (3.7) 5 (1.9) 57 (4.3)

7


Officer/Clinical Officer Midwife 50 (40.0) 137 (60.9) 329 (46.5) 159 (61.4) 675 (51.3) TBA 2 (1.6) 1 (0.4) 1 (0.1) 0 (0) 4 (0.3) No data1 42 (33.6) 65 (28.9) 321 (45.4) 87 (33.6) 515 (39.1)

1 No information on attendant at delivery in database

Analyzing the location of delivery by month of the project as presented in Figure 2, we see no significant decrease in either health facility or home birth rates after the project started (March to August 2009).

Figure 2: Location of delivery by month (n=1,316)

3.4 Feasibility: Misoprostol and IEC Message Coverage

3.4.1 Coverage of IEC Messages The project has run a successful community awareness campaign, utilizing radio, print and person-to-person communication (ANC education session and community awareness activities of CORPs and TBAs) strategies to reach women with safe delivery messages. In the postpartum interview, participants were asked if they had received information on PPH and misoprostol, and the sources from which they learned this information. Their responses are presented in Table 4.

The campaign was also effective in reaching women, with 95% of respondents stating that they had received information on PPH. Of note, health providers and facilities were the most frequently cited sources of information about bleeding after childbirth (83%). Posters/pamphlets (29%), friend/relative (16%), community health worker (16%), radio (15%), and TBA (14%) were mentioned less frequently as sources of information about PPH.

In addition to the information they received on PPH, almost all participants received information about misoprostol, 99% of respondents. All respondents received misoprostol information from a midwife and/or health facility. Other sources of misoprostol information were similar to those noted in PPH information sources. There was some variation across districts in reach of the different communication methods. More participants reported learning about PPH and misoprostol from the radio in Kilombero and Ulanga (35% and 29% respectively vs. 5% in Kigoma and Rufiji). TBAs played a minor role in reaching women with the campaign messages in all districts except Rufiji (15% vs. less

8


than 10% in the other districts). Posters and pamphlets were reported much less often in Kigoma (14% vs. 29% to 34% in the other districts), most likely due to the fact that this district did not receive clinic or community posters until later in the project timeline.

Table 4: Community outreach exposure (n=1,316) Kigoma

n=125 Kilombero

n=225 Rufiji

n=707 Ulanga n=259

Total n=1,316

Received information about bleeding after childbirth

99 (79.2) 224 (99.6) 672 (95.1) 255 (98.5) 1,250 (95.0)

Main sources of information on bleeding after childbirth

Midwife/Health Facility 79 (63.2) 213 (94.7) 565 (79.9)

229 (88.4)

1086 (82.5)

Community Health Worker

9 (7.2) 30 (13.3) 122 (17.3)

43 (16.6) 204 (15.5)

TBA 6 (4.8) 15 (6.7) 137 (19.4)

23 (8.9) 181 (13.8)

Friend/Relative 4 (3.2) 20 (8.9) 126 (17.8)

62 (23.9) 212 (16.1)

Radio 6 (4.8) 78 (34.7) 36 (5.1) 75 (29.0) 195 (14.8)

Posters/Pamphlets 17 (13.6) 65 (28.9) 212 (30.0)

85 (33.8) 379 (28.8)

Received information about misoprostol

120 (96.0) 222 (98.7) 698 (98.7) 256 (98.8) 1,298 (98.5)

Main sources of information about misoprostol

Midwife/Health Facility 109 (100) 221 (100) 689 (100) 251 (100) 1,270 (100)

Community Health Worker

38 (30.4) 33 (14.7) 128 (18.1)

40 (15.4) 239 (18.2)

TBA 6 (4.8) 11 (4.9) 148 (20.9)

19 (7.3) 184 (14.0)

Drug vendor/Pharmacist 0 (0) 2 (0.9) 1 (0.1) 14 (5.4) 17 (1.3) Friend/Relative 4 (3.2) 31 (13.8) 92 (13.0) 71 (27.4) 198

(15.1) Radio 7 (5.6) 67 (29.8) 28 (4.0) 89 (34.4) 191

(14.5) Posters/Pamphlets 18 (14.4) 67 (29.8) 187

(26.5) 79 (30.5) 351

(26.7)

As seen in Figure 3, midwife/health facility was mentioned as the most important source of misoprostol information across the four districts (78% to 95%), with the other sources listed in very small numbers. In Kigoma district, community health workers played an important role in misoprostol information dissemination that was not as profound as in other districts (20% vs. 1 or 2% elsewhere).

9


Figure 3: Most important source of misoprostol information by district (n=1,316)

3.4.2 Coverage of Misoprostol Distribution ANC providers were quite effective in recruiting women into the project, enrolling 92% of women who came to ANC into the project in this preliminary analysis. However, as seen in Table 5, misoprostol was only distributed to about half of the women enrolled in the project; ranging from a low of 41% in Ulanga to a high of 72% in Kigoma.

Table 5: Coverage of misoprostol distribution to enrolled clients at ANC visits (n=4,829) Kigoma Kilombero Rufiji Ulanga Total Total ANC attendance 845 1,380 1,688 916 4,829 Enrolled in the project 830 (98.2) 1,253 (90.8) 1,522 (90.2) 834 (91.1) 4,439 (91.9) Misoprostol distribution 610 (72.2) 641 (46.4) 904 (53.6) 374 (40.8) 2,529 (52.4)

The gap between high enrollment and lower misoprostol distribution is seen in Figure 4. Each month, enrollment nearly matches ANC attendance, meaning that from the beginning of the project ANC providers were able to recruit women coming to ANC into the project. However, ANC attendance at 32 weeks gestation or greater – when women are eligible to receive misoprostol tablets – is much lower than the overall ANC attendance rate. Misoprostol was distributed to almost all women who were enrolled in the project and attended ANC after 32 weeks gestation, meaning that ANC providers were properly distributing misoprostol to eligible women. Therefore, while the ability of ANC providers to recruit women into the project is demonstrated, their capacity to distribute misoprostol to women is inhibited by the gestational age requirement.

10


Figure 4: ANC attendance, enrollment, ANC attendance at 32 weeks gestation, and misoprostol distribution (n= 4,829)

While we anticipated that all women would return for an ANC visit at least once after they reached 32 weeks gestation (i.e. when they were eligible to receive misoprostol), we see in Figure 5 that this was true of only 54% of enrolled participants. Of note, there was wide variation between districts, ranging from 41% in Ulanga to 72% in Kigoma.

Figure 5: ANC attendance at 32 weeks or greater gestation

3.5 Program Effectiveness: Comprehension of IEC Messages and Coverage of Misoprostol at Home Births

3.5.1 Comprehension of IEC Messages Overall comprehension of IEC messages was high amongst participants in the postpartum interview sample (Table 6). When asked during the postpartum interview what information they received

11


about excessive bleeding, most respondents said that it can cause death (87%) and recognized that excessive bleeding occurs when two or more kangas are soaked with blood (81%).

The community awareness campaign, in particular the facility-based education sessions, was highly effective in educating women about misoprostol as comprehension of the key misoprostol messages was very high amongst respondents. Almost all respondents in the postpartum interview knew the function (98%), correct timing (98%), and correct route (99%) of misoprostol for PPH prevention. Additionally, 70% of participants knew of at least one symptom of misoprostol use, with shivering (60%) and nausea (58%) noted most frequently.

Table 6: Comprehension of IEC messages (n=1,316) Kigoma

n=125 Kilombero

n=225 Rufiji

n=707 Ulanga n=259

Total n=1,316

Information about PPH PPH can cause death

96 (76.8) 195 (86.7) 608 (86.0) 249 (96.1) 1,148 (87.2)

Go to a health facility or get help from a midwife 24 (19.2) 121 (53.8) 286 (40.5) 106 (40.9) 537 (40.8) Bleeding soaks 2 kangas or more 90 (72.0) 180 (80.0) 577 (81.6) 212 (81.9) 1,059 (80.5)

Information about misoprostol Misoprostol prevents, stops, or reduces the chances of bleeding after childbirth 120 (96.0) 220 (97.8) 697 (98.6) 255 (98.5)

1,292 (98.2)

Correct timing (immediately after baby is born) 120 (96.0) 220 (97.8) 693 (98.0) 256 (98.8)

1,289 (98.0)

Correct dose (3 tablets) 120 (96.0) 221 (98.2) 698 (98.7) 255 (98.5)

1,294 (98.3)

Correct route (oral) 120 (96.0) 222 (98.7) 698 (98.7) 256 (98.8)

1,296 (98.5)

Knows potential symptoms of misoprostol use

Shivering 76 (60.8) 130 (57.8) 439 (62.1) 150 (57.9) 795 (60.4) Nausea 75 (60.0) 116 (51.6) 409 (57.9) 159 (61.4) 759 (57.7) Vomiting 56 (44.8) 92 (40.9) 346 (48.9) 131 (50.6) 625 (47.5) Diarrhea 39 (31.2) 70 (31.1) 317 (44.8) 105 (40.5) 531 (40.4) Raise of temperature 15 (12.0) 29 (12.9) 196 (27.7) 59 (22.78) 299 (22.7) At least one potential symptom 80 (64.0) 140 (62.2) 496 (70.2) 191 (73.8) 907 (68.9)

3.5.2 Use of Misoprostol at Home Births and Protected Births Key to the program effectiveness of this project is reaching women who deliver at home with misoprostol during ANC. Of the 508 home deliveries (34%) reported amongst the postpartum interview respondents, 93% received misoprostol at an ANC visit. All four districts were quite successful in reaching women who deliver at home with the drug, with over 90% of women reporting receiving misoprostol at ANC. Women who delivered ultimately delivered at a health facility were less likely to have taken misoprostol home from ANC (70%).

12


Of those who received misoprostol at ANC and delivered at home (n=473), almost all used misoprostol at delivery (97%). Therefore, to date the project has achieved relatively high coverage of women who delivered at home with misoprostol distributed at ANC, and very high use of the drug of those who received misoprostol during ANC.

Table 7: Coverage of misoprostol at home births Kigoma Kilombero Rufiji Ulanga Total Home birth1 43 62 316 87 508

Misoprostol distribution at ANC Took misoprostol home 42 (97.7) 57 (91.9) 294

(93.0) 80 (92.0) 473 (93.1)

Did not take misoprostol home 1 (2.3) 5 (8.1) 22 (7.0) 7 (8.0) 35 (6.9) Uterotonic at delivery

Misoprostol 40 (93.0) 55 (88.7) 289 (91.5)

79 (90.8) 463 (91.1)

No uterotonic 3 (7.0) 7 (11.3) 27 (8.5) 7 (9.2) 45 (8.9) Facility birth 82 163 391 172 808

Misoprostol distribution at ANC Took misoprostol home 70 (85.4) 114 (69.9) 284

(72.6) 95 (55.2) 563 (69.7)

Did not take misoprostol home 12 (14.6) 49 (30.1) 107 (27.4)

77 (44.8) 245 (30.3)

Uterotonic at delivery Misoprostol 48 (58.5) 57 (35.0) 24 (6.1) 107

(62.2) 236 (29.2)

Injection 28 (34.2) 84 (51.5) 331 (84.7)

48 (27.9) 491 (60.8)

No uterotonic 6 (7.3) 22 (13.5) 36 (9.2) 17 (9.9) 81 (10.0) 1Includes births en route to the health facility

In addition, overall coverage of deliveries with an uterotonic for PPH prevention after delivery within the district and proportion of births protected against PPH (“protected births”) are important indicators that the project is meeting its goals and objectives. Overall coverage of deliveries with an uterotonic at the district level was high. Most participants received an injection at a health facility (37%), or misoprostol at a health facility (18%) or at home (35%). There was some variation between districts on the uterotonic used. Misoprostol use at a home delivery ranged from 31% in Ulanga to 41% in Rufiji. Use of injection at delivery for PPH prevention was highest in Rufiji (47%) and lowest in Ulanga (19%). Misoprostol use at health facilities varied greatly, from 3% in Rufiji to 41% in Ulanga.

Table 8: Coverage of protected births (n=1,316) Kigoma

N=125 Kilombero

N=225 Rufiji

N=707 Ulanga N=259

Total N=1,316

Uterotonic for PPH prevention

Injection 28 (22.4) 84 (37.3) 331 (46.8) 48 (18.5) 491 (37.3) Misoprostol at health facility 47 (37.6) 56 (24.9) 23 (3.3) 107 (41.3) 233 (17.7) Misoprostol at home 41 (32.8) 56 (24.9) 290 (41.0) 79 (30.5) 466 (35.4) No uterotonic 9 (7.2) 29 (12.9) 63 (8.9) 25 (9.7) 126 (9.6)

Births protected from PPH1 116 (92.8) 196 (87.1) 644 (91.1) 234 (90.4) 1,190 13


(90.4) 1 Any uterotonic given for PPH prevention

A protected birth for the purposes of this analysis is a delivery where the mother is protected from PPH by use of any uterotonic administered or taken for PPH prevention. Of those who participated in the postpartum interview, use of any uterotonic after delivery was quite high at 90% (see Figure 6. Slightly more than half of these protected births were at a health facility, with either an injection (37%) or misoprostol (18%). Misoprostol at home deliveries protected an additional 35% of deliveries that occurred at home, deliveries that would not have been protected were it not for misoprostol distribution at ANC. Based on a previous study conducted in Kigoma (Prata, Mbaruku et al. 2005), the incidence rate for PPH (bleeding of 500mL or more) is estimated to be around 20%. Using this as a measure, we estimate that of the 699 women who did not receive an injection to prevent PPH, around 140 women would develop bleeding after delivery that would necessitate referral and/or additional interventions to stop the bleeding.

As seen in Figure 7, protected births at home were high at 91%. This demonstrates the program’s effectiveness and impact; without misoprostol these women would have been unprotected from PPH.

Figure 6: Coverage of uterotonic for PPH prevention (n=1,316)

Figure 7: Coverage of protected births at home vs. health facility deliveries (n=1,316)

3.5.3 Bleeding-related referrals and the need for additional interventions Overall, bleeding-related referrals were low amongst women who participated in the postpartum interview; only 11 women reported requiring referral for bleeding-related causes (1%). A bleeding-related referral includes both excessive bleeding and retained placenta, and is based on the participant’s self-report in the postpartum interview. Figure 8 shows the distribution of bleeding-related referrals and need for additional interventions by location of delivery and use of uterotonic after delivery.

14


Of the 463 women who delivered at home and took misoprostol, only three were referred for bleeding-related reasons (1%) and two required additional interventions. One woman who delivered at home and did not use misoprostol required both a referral for bleeding-related reasons and additional interventions.

Of the 808 women who delivered at a health facility, most received an uterotonic after delivery (90%) and only 1% of these women required referral and additional interventions. A small proportion of women who delivered in a health facility reported not receiving an uterotonic after delivery (10%), and only one woman reported requiring referral and additional interventions for bleeding-related reasons (1%).

Figure 8: Bleeding-related referrals and need for additional interventions

3.6 Safety: Correct Use of Misoprostol and Postpartum Symptoms Of women who used misoprostol at home births, all reported using the correct dose and route of misoprostol. Therefore, correct misoprostol use was universal for all participants who used the drug at a home delivery for PPH prevention.

Table 10: Correct use of misoprostol at home births (n=463) Kigoma

N=40 Kilombero

N=55 Rufiji

N=289 Ulanga

N=79 Total

N=463 Correct dose (3 tablets) 40 (100) 55 (100) 289 (100) 79 (100) 463 (100) Correct route (oral) 40 (100) 55 (100) 289 (100) 79 (100) 463 (100) Correct use of misoprostol (correct dose and route) 40 (100) 55 (100) 289 (100) 79 (100) 463 (100)

Participants’ self-report of symptoms experienced after delivery is presented in Table 11. Almost four out of five women did not experience any symptoms during the postpartum period (79%). Experience of symptoms was higher amongst misoprostol-users (26% vs. 16% of non-users;

15


p<0.001). The most common symptoms experienced were shivering (16%), nausea (6%), and raise of body temperature (5%). Misoprostol users were significantly more likely to report experience of these three symptoms. Report of multiple symptoms was low (6%), and more common amongst misoprostol-users (8% of those who reported symptoms, vs. 4% of non-users; p=0.001). Symptoms self-resolved within an hour for almost all participants.

Table 11: Reported experience of postpartum symptoms (n=1,316) Took

misoprostol1 (n=699)

Did not take misoprostol2

(n=617)

Total

(n=1,316) p-value Did not experience postpartum symptoms 187 (26.8) 95 (15.6) 283 (21.5) <0.001 Experienced postpartum symptoms 521 (84.4) 512 (73.3) 1,033 (78.5) <0.001

Shivering 142 (20.3) 65 (10.5) 207 (15.7) <0.001 Nausea 57 (8.2) 23 (3.7) 80 (6.1) 0.001 Vomiting 20 (2.9) 16 (2.6) 36 (2.7) 0.766 Raise of body temperature 44 (6.3) 23 (3.7) 67 (5.1) 0.035 More than one symptom (range 2 to 4) 57 (8.2) 24 (3.9) 81 (6.2) 0.001

1 At home or facility birth 2 Includes women who received injection (n=491)

3.7 Acceptability: User Perspectives on Misoprostol Use Acceptability of misoprostol was high amongst users and non-users alike. Almost all women would recommend misoprostol to a friend (98%), use misoprostol in a subsequent pregnancy (97%), or purchase misoprostol (94%). While women who used misoprostol were significantly more likely to answer positively to the acceptability questions, the difference was quite small and in none of the acceptability indicators did non-users respond less than 90%.

Table 12: Acceptability of misoprostol Took

misoprostol1

(n=699)

Did not take misoprostol2

(n=617)

Total

(n=1,316) p-value Would recommend misoprostol to a friend 693 (99.1) 593 (96.1)

1,286 (97.7) 0.001

Would use misoprostol in a subsequent pregnancy 690 (98.7) 586 (95.0)

1,276 (97.0) <0.001

Would purchase misoprostol 669 (95.7) 564 (91.4)

1,233 (93.7) 0.001

Average amount willing to pay for misoprostol in Tanzania Shillings (+ SE)

1598.6 + 44.8

1842.1 + 69.9

1709.9 + 40.3 0.003

1 At home or facility birth

4. Conclusions This report presents preliminary results of the joint operations research project to assess the feasibility, program effectiveness, safety, and acceptability of misoprostol distribution via ANC visits to prevent PPH at home births.

16


Overall, these preliminary results are positive. In terms of feasibility, ANC providers have been successful in recruiting women to the project during ANC visits, and we are seeing more women enroll in the project than expected. In addition, of those who are eligible to receive misoprostol, virtually all women are given the drug during an ANC visit after they have reached 32 weeks gestation. One challenge illuminated in these data is that women are returning to ANC after 32 weeks at a rate that is lower than expected (54%).

Feasibility in terms of reach of the community awareness campaign is also showing promising results. Most participants analyzed thus far have received information about PPH and/or misoprostol. The variation in reach of the different communication vehicles is most likely explained by differences in implementation of the campaign at the district level.

Health facilities and providers based in facilities were the most cited and most important sources of information on PPH and misoprostol. This is not surprising given that all women should have been presented with the Education Session during ANC. However, the importance that participants placed on the health facility for this information demonstrates the vital role health facilities and their staff play in educating women. This suggests that moving forward the facility-based education sessions may be the critical IEC component and should be prioritized over other communication elements of the program.

Preliminary results on the program effectiveness in terms of coverage of home births with misoprostol and coverage of protected births are very promising. Of those who delivered at home, the majority (93%) had received misoprostol during ANC, and almost all of these women took the drug after delivery (97%). All women who used misoprostol at home delivery used the drug correctly. The high use rate and correct use of misoprostol is vital to the effectiveness of the project, and also demonstrates the effectiveness of the awareness campaign. As noted in the analysis, knowledge of PPH, its consequences, and the use of misoprostol was high amongst respondents. This most likely played an important role in women choosing to use the drug after a home birth, and using it correctly.

In addition, protected births were promisingly high – 90% of postpartum interview respondents were protected from PPH by use of an uterotonic after delivery. Misoprostol plays a vital role in protecting births both at home and in health facilities when other uterotonics are not available. During monitoring visits for the project, it was noted that prior to the project, dispensaries were not stocked with an uterotonic for PPH prevention after delivery. Once the project began, participants began returning to the dispensary for delivery and bringing the misoprostol that had been distributed to them during ANC, as they had been told to do during the Education Session. Thus, availability of misoprostol has increased uterotonic coverage at the dispensary level and a large proportion of the 18% of protected births using misoprostol at a health facility may be due to this reason. Lastly, through this project an additional 35% of births were protected using misoprostol at home deliveries after being distributed during ANC, vastly increasing the proportion of protected births in the population.

Of note, misoprostol distribution at ANC had no effect on the proportion of health facility deliveries of those who participated in the postpartum interview.

The safety of misoprostol distributed at ANC for PPH prevention at home births was well demonstrated. All women who used the drug at home used it correctly, taking the correct dose via

17


the correct route. Furthermore, experience of symptoms was quite infrequent, although more frequent compared to those who did not use the drug. These symptoms were self-resolving within an hour for most cases, and none required referral.

Misoprostol is very acceptable to women. Despite reported symptoms after use, almost all women would recommend misoprostol to a friend, use it again in a subsequent delivery, or purchase misoprostol. Therefore, it can be assumed that the symptoms related to misoprostol use are of minor consequence compared to the perceived benefit of the drug.

While the results of this preliminary analysis are encouraging, it is important to mention the limitations of this analysis. These results are based on analysis of data available at the time of writing. The project is still ongoing and postpartum interviews are still being conducted. Once the project is complete, including postpartum interviews of participants, a full analysis will be conducted. Since there will be more data, there is potential for some of the results and trends seen in this analysis to change when the final analysis is conducted.

5. Recommendations Given the positive findings of this preliminary analysis, we recommend to policy makers and key stakeholders to begin planning for scale-up of misoprostol distribution at ANC nationwide in Tanzania.

The findings in this preliminary report warrant further analysis. In particular, we want to look further into the gap between enrollment and misoprostol distribution, especially given that a large proportion of women delivering at home reported that they received misoprostol during ANC. After all the data has been entered, we will write a more complete report to present the final results of the project.

ANC attendance after 32 weeks gestation when women are eligible to receive misoprostol was lower than expected. Therefore, we recommend greater effort to encourage women to return to ANC after they have reached the gestational age requirement in order to receive the drug. However, It is possible that women’s tendency to not attend ANC later in pregnancy may be due to logistical factors rather than a lack of desire to return to ANC, and therefore lowering or removing the gestational age limit should be considered so that more women attending ANC can be eligible to take misoprostol home with them.

18


6. References Alfirevic, Z., J. Blum, et al. (2007). "Prevention of postpartum hemorrhage with misoprostol."

International Journal of Gynecology & Obstetrics 99(Supplement 2): S198-S201.

Derman, R., B. Kodkany, et al. (2006). "Oral misoprostol in preventing postpartum haemorrhage in resource-poor communities: a randomised controlled trial." The Lancet 368(9543): 1248.

Geller, S. E., M. G. Adams, et al. (2006). "Postpartum hemorrhage in resource-poor settings." International Journal of Gynecology & Obstetrics 92(3): 202-211.

Joy, S. D., L. Sanchez-Ramos, et al. (2003). "Misoprostol use during the third stage of labor." International Journal of Gynecology & Obstetrics 82(2): 143-152.

Lagenbach, C. (2006). "Misoprostol in preventing postpartum hemorrhage: A meta-analysis." International Journal of Gynecology & Obstetrics 92: 10-18.

Maine, D. (1993). Safe motherhood programs: options and issues. New York, Columbia University.

NBS Tanzania and ORC Macro (2005). Tanzania Demographic and Health Survey 2004-05. Dar es Salaam, Tanzania, National Bureau of Statistics [Tanzania] and ORC Macro.

Prata, N., A. Gessessew, et al. (2009). "Prevention of postpartum hemorrhage: Options for home births in rural Ethiopia." African Journal of Reproductive Health 13(2).

Prata, N., G. Mbaruku, et al. (2005). "Controlling postpartum hemorrhage after home births in Tanzania." International Journal of Gynecology & Obstetrics 90: 51-55.

Tang, O., D. Gemzell-Danielsson, et al. (2007). "Misoprostol: Pharmacokinetic profiles, effects on the uterus and side-effects." International Journal of Gynecology & Obstetrics 99: S160-167.

Walraven, G., J. Blum, et al. (2005). "Misoprostol in the management of the third stage of labour in the home delivery setting in rural Gambia: a randomised controlled trial." British Journal of Obstetrics and Gynaecology 112(9): 1277.

WHO (2005). Maternal Mortality in 2005: Estimates developed by WHO, UNICEF, UNFPA, and the World Bank. Geneva, The World Health Organization.

19

misoprostol distribution during antenatal care visits ... · pdf filemisoprostol distribution...

Documents