Home / Maama project

Maama project

Evaluation  of the Maama  project:

Evaluation  report 2016

  Abbreviations   3   3.2 CHW Home visits   15
Executive summary 4 3.3 Maama kits 17
Summary of recommendations 5 3.4 Antenatal care attendance – the women’s perspective 19
1. Introduction 6 3.5 Antenatal care visits – the clinic perspective 20
1.1 Context 7 3.6 Deliveries 22
1.2 Maternal and newborn health in Uganda 7 a) Delivery location and birth attendance 22
1.3 The Maama Project 7 b) Birth preparedness and transportation 23
2. Evaluation methodology 9 c) The role of traditional birth attendants (TBAs) in delivery 23
2.1 Aim and objectives 10 3.7 Newborn health 25
2.2 Evaluation design, tools and data collection 10 a) Newborn care practices 25
2.3 Study population and sampling strategy 11 b) Umbilical cord care 25
2.4 Ethics 11 2.8 Male involvement 26
2.5 Data analysis 11 3.9 Mental health 28
3. Findings 12 3.10 Family planning 28
3.1 Characteristics of project beneficiaries 13 References 31

Core evaluation team: Linn Persson Berg, Erika Lejon Flodin, Hedvig Berntell, Teresa Marie Kreusch, Tania Neuman, Evelina Linnros, Amrita Namasivayam and Marjan Molemans

Graphic designer: César Augusto Ortelan Perri

ANC Antenatal care

CHW Community health worker

NGO Non-governmental Organisation

SOGH- Swedish Organization for Global Health

TBA Traditional birth attendant

WHO World Health Organization

Executive summary

This report documents the evaluation  of the Maama project following two years of implementation. The project aims at increasing knowledge of maternal  and newborn  health  in the community  and the uptake  of health  services such  as antenatal care visits and  deliveries at health facilities.

The project is a collaboration between Swedish Organization for Global

Health (SOGH) and Uganda Development and Health Associates (UDHA).

The CHWs conducted 1231 home visits during the project year. Of the interviewed women, 80% reported getting at least one prenatal home visit and 84% of women received at least one postnatal visit.

The proportion of women completing four ANC visits increased from 12%

at baseline to 86% and health facility deliveries increased from 70% to

89%.

A total of 246 birth kits were distributed during ANC visits. Out of the women interviewed, 79% reported receiving a birth kit.

The assessment  indicates that  the birth kit is a strong motivator for pregnant women to complete four antenatal visits and to give birth at a health facility by addressing financial barriers.

The CHWs and the clinic staff have reported an overall empowerment of the community and a decrease in misconceptions and knowledge gaps.

The community is aware of health issues and the health seeking behavior has increased in the community.

Of the mothers who had received chlorhexidine, 96% used it and 81% used in within 24 hours. All interviewed women had a positive attitude towards the use of chlorhexidine for umbilical cord care.

4

Summary of recommendations

Already implemented

The need  to improve  the delivery of kits to the clinic was met during summer  2016 in two ways. First, instead  of ordering  kits a couple  of times per year and waiting for them to be delivered, the supply of kits for a whole year was ordered at once. Secondly, vouchers were made in case the stock does run out, so that mothers  can come and collect the kit later on.

The limited capacity of Maina clinic was remedied during the evaluation by the purchase of an extra bed where women who have delivered can take rest before making the journey home.

The  CHW have  received  extra  training  to  be  explain  the  use  and application of chlorhexidine to ensure better use in the future. The Maina clinic also has a pictorial on the wall explaining how it should be used.

Short term

SOGH should look at the possibilities to help the CHW create a better system to identify pregnant woman and give additional aid, for example in form of a card or booklet, to help them remember all topics that need to be covered in the home visits.

Investigate the  possibility to include more  items in the  kit, especially more plastic gloves. However this is difficult, since the kits are bought in a sealed bag. Another way to accommodate for the lack of gloves would be to supply the clinic with extra gloves.

Make the  conditions  for receiving  a  Maama  kit more  clear  for the beneficiaries (i.e. four ANC visits at Maina clinic)

CHWs should encourage the women  in the villages to go to the ANC clinic earlier during their pregnancy, so they will have time to complete all four visits.

Educate CHW more on family planning and encourage them to pass on the knowledge.

Educate husbands on family planning

Distribution of a mother’s card with the important information  about the women’s ANC visits. The card could also be used as a check list at the ANC visits, but also at the home visits by the CHW.

Continue  to  spread  of information  in the  communities,  primarily in the  rural areas, with an emphasis  to address  the  importance of men in  understanding and  taking  full responsibility  as  fathers,  through community sensitization taking place monthly.

Long term

Donation of bikes to the CHW to facilitate the home visits. Increased compensation for the CHW

A larger facility with more seating areas and at least one more antenatal/

delivery bed and installation of electricity and running water.

The workforce should also be increased, preferably with a midwife. Provision of items that are currently lacking should be provided to the

clinic, e.g. blood pressure machine and a measuring tape.

Extra stock of medicine and rapid malaria tests. Adding rapid HIV, syphilis tests and more options of contraceptives to the stock of medicine.

Strengthen and support the group of single mothers.

In the long term, it should be considered  if SOGH can organise a system to make it more easy for mothers  to get  to the clinic. For example, a bodaboda that mothers in need can call.

1.Introduction

6

1.1  Context

Every year 2.7 million infants die during their first month  of life (1). An estimated 99% of these  deaths  take place in low- and middle-income countries  (2). During the past two decades, neonatal  death  rates have out of all low- and middle-income  regions decreased the least in sub- Saharan Africa (3): 29 neonatal  deaths  occur for every 1000 live births in the region, compared to 3 per 1000 in high-income  countries (1). At the current  rate of change, it will be over a century before an African newborn  has the same chance of survival as a baby born in Europe or North America (4), indicating a pressing need for interventions targeting pregnancy, childbirth and the newborn period.

Newborn health and survival are closely related to maternal health. Over half of all maternal  deaths  in the world are due to preventable causes such as hemorrhage, hypertensive  disorders  and sepsis. These can be addressed with quality antenatal, delivery and postnatal care, including, completion  of four antenatal visits, skilled care  during  delivery, and postnatal visits during the first week after birth (5). Research has shown that  most  newborn  deaths  can  be  prevented with already  available interventions  targeting  preconception,  antenatal,  intrapartum and postnatal care  (6). Skilled care  during  labor  is estimated to  reduce neonatal deaths by 25% and a combination of clean birth and postnatal care practices can reduce neonatal  deaths due to sepsis and tetanus by

40% (6). Community-based care that includes community mobilization, home visits and improved linkage to health care services has also been estimated to reduce neonatal  mortality by 40% (6).

1.2 Maternal and newborn health in Uganda

Even though Uganda achieved the Millennium Development Goal 4 by reducing  under-five mortality to less than  90 per 1000 live births, the neonatal  mortality rate still remains high at 19/1000 births. Over one- third (35%) of under-five deaths happen during the first month of

life. Uganda  fell short  of achieving  the  desired  5.5% reduction  in the maternal mortality rate of Millennium goal 5. Moreover, the lifetime risk of death due to pregnancy or childbirth remains high at 1 in 44, with 343 maternal deaths per 100,000 live births (5). The health care coverage of essential interventions remains overall low in the country. Only 44% of the women meet with the demand for family planning, only 48% of the women  complete  four ANC visits and only 57% of the women  have a skilled attendant present  at delivery. Moreover, only 33% of the women receive postnatal care and  63% of women  breastfeed exclusively for the first six months  (5). There is an equity gap between the richest and poorest, with large differences in an unmet  need  for family planning, completion  of four  ANC visits and  presence   of skilled attendant at delivery (5).

Research conducted at the Iganga-Mayuge  Demographic  Surveillance Site in southeastern Uganda has contributed significantly to the knowledge on the state of newborn health in rural Uganda. The research indicates  that  that  54% of newborn  deaths  occur away from a health facility and half of all newborn deaths are linked to a delay in the decision to seek care. Most newborn  deaths  in the area happen during the first week of life: 47% during the first 24 hours, and 78% during the first seven days (7). An additional challenge is the lack of knowledge regarding safe newborn care practices, with coverage ranging from 38% for clean cord care, 42% for optimal thermal care and 57% for exclusive breastfeeding (8).

1.3 The Maama  Project

To address  the  risks mothers  and  newborns  face  in Uganda,  SOGH developed  a  maternal   and  newborn   health   project  together  with Uganda  Development  and  Health  Associates  (UDHA).  The  Maama Project covers Maina Parish, located in Mayuge District in southeastern

Uganda. The project area consists of five villages (Mwezi, Kyete, Maina, Girigiri and Bulondo) and a private health center financed by the partner NGO UDHA. The health  center  is classified as level II,  denoting the most basic level of facility health  care out of four possible levels, with village health teams comprising level I. The Maama Project follows the recommendations outlined  in a joint statement by WHO and UNICEF that recommends the uptake of a home visit strategy to reduce newborn deaths (9). The project is based on a community model of two prenatal and three postnatal home visits that has been tested  and evaluated  by several studies (10–16), including the Uganda Newborn Study (UNEST) conducted in Iganga and Mayuge districts (17,18).

Project activities are carried out by CHWs who have been picked out by local leaders and trained by staff from the Iganga-Mayuge Demographic Surveillance Site. CHW’s main role is to identify pregnant women  and provide two prenatal and three postnatal home visits (on days 1, 3 and

7 after birth).

During postnatal visits, the CHW counsels the woman on safe newborn care practices and family planning. The CHWs have also been trained to identify low birth weight babies and provide referrals to health facilities.

Furthermore,  to  promote  hygienic  practices  during  deliveries,  the project includes the provision of birth kits (Maama kits), which are pre- prepared, packaged, single-use  kits that  contain  a selection  of items pivotal to a hygienic delivery. The kit contains two pairs of sterile gloves, cotton  wool,  sterile blade,  a preparation sheet,  a plastic sheet,  soap, cord tires and a new child growth and postnatal clinic card. The birth kits are provided  by the health  facility to pregnant women on their fourth antenatal visit, acting as an incentive for the women to attend ANC four times.

In summary, a pregnant woman in the project area receives two home visits from a CHW and  visits a health  facility four times  during  her pregnancy. On the fourth visit she receives a Maama kit, which can be used during a facility- or home delivery. After delivery, she receives three postnatal home visits from the CHW.

2. Evaluation  methodology

9

2.1 Aim and objectives

The aim of the  evaluation  was to assess to which extent  the  Maama Project has been successful in improving maternal and newborn health in Maina Parish.

The main objective of the evaluation was to describe the change in the community  regarding  attitudes and behaviors related to maternal  and newborn  health  following two years of implementation. A secondary objective  was  to  identify  challenges   of  the  project  and  remaining barriers related to health service uptake, as well as ways to develop and expand the project.

The qualitative and quantitative key indicators investigated were: Completion of four ANC visits

Delivery with a skilled birth attendant

Uptake and knowledge of safe newborn care practices

Number and timing of home visits

Services and education provided during the home visits

Services and education provided during visits at the Maina clinic

The role and use of Maama kits

The role of community health workers in attitude and behavior change

Perceptions of the successes and challenges of the project

The mental health of the women, during and after pregnancy

Male involvement in maternal health, newborn care and antenatal care

The use of Chlorhexidine, a disinfectant for the umbilical cord stump

Due to the relatively small number of estimated pregnancies and deliveries in the project area and the small sample size in this evaluation, any effect of the project on maternal or newborn mortality, which are rare outcomes in themselves, could not be investigated.

2.2 Evaluation design, tools and data collection

A mixed-methods design  was used  to assess the  effectiveness  of the project. During the evaluation period, both quantitative and qualitative data  were collected  through  interviews and a surveys. Data collected at baseline  and during  the project  year (2015-2016) were included  in the  analysis. The interview  tools  were  developed in English by the evaluation   team   consisting   of  eight   SOGH interns.  Three  Lusoga- speaking  interpreters  worked  in  the  evaluation.  While two  female interpreters who  were  not  involved  in  the  project  interpreted the majority  of  interviews,  the  male  project  manager   interpreted the interviews on male involvement, one interview with a CHW and three interviews  with mothers  during  one  field day. The evaluation  group made this decision because  it was suggested that the husbands would feel more comfortable with a male interpreter.

Beneficiary interviews served as the main source of quantitative data. A modified questionnaire based on the Demographic Health Survey (1) was used to interview program beneficiaries, i.e. mothers in the project area. The final questionnaire consisted  of the  following components: respondent’s background, reproduction, pregnancy  and postnatal care, use of chlorhexidine, contraception, occupation and family economy. Additional questions  concerning  antenatal care attendance, CHW visits and birth kit use were included. In addition, data from monitoring tools that were filled in monthly by CHWs and clinic staff from August 2015 to July 2016 have been included in the quantitative analysis.

Qualitative  data  on  newborn   and  maternal   health   were  obtained through  semi-structured interviews  with CHWs and  health  care staff from Mayuge health  centre  and  Maina clinic. In addition,  qualitative interviews with mothers  and fathers participating  in the project during the last year and traditional birth attendants (TBAs) were conducted. All interviews were recorded and transcribed.

2.3 Study population and sampling strategy

Inclusion criteria for the  mothers  for both  qualitative  interviews and quantitative surveys were i) living in the project area, ii) having given birth during the past 12 months  and iii) being available for interviews during  the  evaluation  period. Random sampling  in the  form of pre- selecting households based  on geographical location was trialled, but turned  out to be impossible  to implement  in the  local setting, partly because many women worked outside the home and could not be found with this method. Participants were thus identified through convenience sampling. Some were approached with the help of CHWs, who located mothers willing to be interviewed. Others were interviewed during their visit to the weekly immunisation  day at the Maina Clinic. A total of 70 mothers  were interviewed, nine of which were excluded in the analysis due to their children being older than 1 year. The final sample size was

61 women.

Five further  parties  were  interviewed   qualitatively.  These  included eleven CHWs from the five villages, the two nursing assistants working at the Maina Clinic, one midwife from the nearby Mayuge Health Centre, one traditional birth attendant (TBA) and four husbands of women  in the Maama project.

2.4 Ethics

Informed consent was obtained orally before commencing the interview. The  participants   were  assured  that  i) their  answers  would  remain anonymous ii) that  they had  the  right to refuse the  interview, refuse to answer specific questions  or stop at any time without  providing an explanation  and iii) that  their responses  would not affect their future health  care. Additionally, permission to audio-record  was obtained for the qualitative interviews.

2.5 Data analysis

The quantitative survey results were entered into a database in SPSS version 23. New summary variables were created  from the data, e.g. a binary ‘completion of 4 ANC visits’ variable. Analyses were performed in  SPSS to  obtain  descriptive  estimates   of  the  outcome variables. Associations and differences between groups were tested  for statistical significance with Chi-square and t-tests.

The qualitative data from interviews with the CHW, health care staff at Maina clinic and Mayuge, TBA and the husbands involved in the project was content analysed to identify patterns. Furthermore, content analysis was made on the data from the interviews with the mothers to provide more  insight  in  specific  topics  covered  in  the  quantitative  survey. Lastly, the qualitative question  items from the quantitative survey were analysed.

3. Findings

12

3.1 Characteristics of project beneficiaries

Table 1 presents the characteristics of the study population. The average age of mothers interviewed was 27.7 years. Close to all women were married and most lived together with their husband. The majority worked as farmers and more than half had completed between 6-10 years of schooling. More

16 18 12 15 39    

than half of the mothers wanted more children, the average desired number of children being 5.5 (SD1 = 1.4) children.

Mean Age at last  birthday

27.7 years

Percentage (%)   Number/ Total N

VILLAGE

Bulondo Giri Giri Kyete Maina Mwezi

MARITAL STATUS

AND LIVING SITUATION

Married/living together with a man as if married

16 18 12 15 39  
87 74 25 8 35 5    
    53/61 45/61 15/61     5/61 18/61     3/61      

Married  and living with husband Married  but not living with husband Separated/divorced

Husband has other wives

Never married

RELIGION

Christian

Muslim

56   34/61 39 24/61 5   3/61    

Other (Born again)

EDUCATION

None

1-5 years

6-10 years

12   7/61 16 10/61 62   38/61 10   6/61    

11 or more

EMPLOYMENT

Farmer

  10/61 11/61 7/61     9/61 24      
75   46/61 3 2/61 21   13/61    

Seller of agricultural products

Other

1. SD = standard  deviation from the mean

Mean Age at last  birthday

27.7 years

Percentage (%)   Number/ Total N

DECISION MAKER FOR HEALTHCARE

Woman herself

Husband/Partner

39 36 13 12    
  24/61 22/61 8/61     7/61      

Woman and partner together

Other

DECISION MAKER

FOR MAJOR HOUSEHOLD PURCHASES

Woman herself

Husband/Partner

10 49 33 8    
  6/61 30/61     20/61     5/61      
4.82 4.52 25 18 20 59 5.5    
  15/61     11/61     12/61     36/61      
    61/61     61/61      

Woman and partner together

Other

    Mean number of pregnancies during  lifetime Mean number of births during  lifetime  
FERTILITY AND FAMILY   Has lost one or more  children in the first month of life Has had a stillbirth, abortion or miscarriage
PLANNING   Currently using a contraceptive method (excl. breastfeeding)
    Wants more  children
    Desired number of children*
MOST RECENT     Most recent birth was a live birth     100    
  BIRTH   Baby from most  recent birth is still alive   100    

Table 1. Characteristics of the respondents of the quantitative survey.

*calculated among  mothers who wanted more  children as current number of living children plus number of desired children.

3.2 CHW Home visits

The project monitoring  tools indicated  that the CHWs conducted 1,231 home visits during the project year 2015-16; each CHW made 103 visits on average. The visits are composed of 862 prenatal visits and 369 postnatal visits. This is more than the 1,021 visits in the previous year, which can partly be explained by the addition of one village to the project area.

Out of the 61 mothers in the evaluation survey, 49 women (80%) received at least one prenatal CHW home visit. That 20% of interviewed women did not receive prenatal visits indicates the need to improve the project’s reach. At the first visit women were on average 4.13 months (SD= 1.65) pregnant, ideally it should occur when being 8-12 weeks pregnant (18). The mean number  of prenatal visits was 3.5 (SD=2.0 visits), which exceeds the desired number of 3 visits.

Eighty-four percent  (51 women) received at least one postnatal home visit from a CHW. The median number  of postnatal visits were 2 (SD= 1,9). The reach thus needs to be improved but the number of visits to those who receive visits meets the desired target.

Forty-seven percent  (29 women) reported that they had been referred to a health facility for something else than antenatal care. The most common reason for being referred was malaria. Other reasons were stomach pain, headache or bleeding, which are possible danger signs for pregnant women. The frequent use of referrals is a positive outcome.

Baby’s foot measured with foot length card

Family planning

28%

49%

50%

72%

Topics covered during  CHW home  visits (N=61)

2014 – 2015

2015 – 2016

Clean cord care Benefits of breastfeeding Thermal care for baby Newborn danger signs

Receipt of Maama kit at 4th ANC visit

Birth preparation

54%

69%

59%

69%

76%

80%

84%

84%

79%

54%

100%

97%

0%         20%     40%      60%      80%      100%     120%

Figure 4 shows the  topics  covered  by the  CHWs during  home  visits, as reported by our sample. Overall, slightly less mothers  reported the different topics to have been covered compared to year 2014-15 (Fig. 1). Last year’s sample was however  more strongly selected  for compliant mothers  than  it is the  case  in the  present  evaluation,  which  might explain the difference. A positive change was observed in the proportion of women  who reported that the CHW had measured the baby’s foot, a method to identify low birth weight. This indicates an improvement thanks to the specific training on how to use the foot length  card that the CHWs received in summer 2015.

Challenges

A part of our sample was obtained through the CHWs leading us to the women  and the CHWs were sometimes  present during the interviews, which may have affected the answers the women gave. No CHWs were present during the interviews conducted at Maina clinic.

There might be cultural factors, such as a tendency towards answering “yes” rather than “no” on questions, that could have an impact on the answers (social desirability bias).

Suggestions

For the next evaluation it might be of value to find another way to ask for what information the mothers received from the CHWs, instead of asking closed yes-or-no questions. In this evaluation they were asked whether or not they remembered being told about this topic. Another way of asking might give more information about the content of what they remembered.

Some  CHW indicated  that  the  workload  was  heavy,  and  that  they sometimes  had to go very far to visit a mother. Therefore, they asked SOGH if it would be possible to donate bikes to them. In the long term, an increased compensation should also be considered.

Some mothers did not receive any home visits because the CHW did not know them or were not aware of the pregnancy. A better system to find pregnant mothers could be considered.

The slight decrease in percentage of topics covered by the CHW, might be due to the heavy workload. An extra aid for the CHW to remember them which topics should be considered, could be useful.

3.3 Maama  kits

The Maama kit is intended to  grant  the  beneficiaries  in the  Maama project the items required for a hygienic delivery. It is distributed to the beneficiaries at the fourth ANC visit and thus works as an incentive for the mothers to seek the recommended amount  of antenatal care.

The Maama kit includes two pairs of sterile gloves, cotton wool, a sterile blade, a preparation sheet, a plastic sheet, soap, cord tires and a new child growth and postnatal clinic card. A total of 246 Maama kits were distributed during the project year 2015-2016. This is an increase with 47 kits compared to the previous year. According to the monitoring  data,

68% of beneficiaries reported using the kit during  delivery. For those who did not use the kit the monitoring data does not inform on whether mothers  had not received it or chose not to use it. Since none  of the interviewed mothers chose not to use the Maama kit, there is reason to assume that it is the first reason.

Out  of the  women  interviewed, 79% (48 women)  reported having received the kit. Among the 13 women  who did not receive a kit, the reported reasons  were  that  (i) ANC visits took  place  at  the  Mayuge district hospital and not at the project  clinic (38%) (ii) four ANC visits were not completed (31%) and (iii) the kit was out of stock at the Maina clinic (23%). One woman completed four ANC visits at the Maina clinic, yet as she was referred to undergo cesarean section at a larger hospital in an early stage of her pregnancy, she did not need the kit.

CHWs and health personnel  alike reported that the Maama kit works as an effective incentive for the women to complete four ANC visits.

“It is easier for the mothers since they receive the Maama kit now. Before it was more difficult for them economically to buy the things they needed for delivery which are in the Maama kit now.”

Peninah, nursing assistant at Maina clinic

At the Mayuge district hospital, the healthcare staff explained that since the hospital is frequently understocked with cotton, clean razor blades, cord ties and other items required for a clean delivery, it is helpful when Maama project beneficiaries bring these items themselves.

Generally, the view among mothers, CHWs and health personnel is that all items in the kit are essential. Some beneficiaries and CHWs reported that the kit should include baby clothes. Healthcare staff at both the Maina project  clinic and the Mayuge district hospital as well as some CHWs, emphasized that  the  kit could  be  improved  by including  additional plastic gloves. Another suggestion from the nursing assistants at Maina clinic was to include gauze pads.

In the in-depth  interviews, the women  who had received a kit agreed with the view expressed by the health personnel and the CHWs, namely that  the kit is an important incentive  for them  to complete  four ANC visits and that it helps them save money.

Furthermore, the  women  expressed  that  the  items  in the  kit were useful, but also stated that the kit could be improved by including more items such as soap, baby clothes, sheets and a towel for the baby. One woman who reported not receiving a kit as she did not complete  four ANC visits, explained that she made one herself by buying the following items: cotton, gloves, clothes, sheets and a razor for the umbilical cord. One interview was conducted with a woman who, due to a quick onset of delivery, was not able to reach the clinic. The family members  who assisted  at her delivery made  use of the  gloves and  the  razor blade, showing how the kit also can be used to make home deliveries safer.

Challenges

The Maama kits sometimes ran out of stock, then women had to come back later to get it or did not receive it. If this happens too often the incentive for completing four ANC visits is taken away.

Mothers and CHWs ask for more items to be included in the kit. However, the Maama kits are bought in a sealed bag and the manufacturer can not provide more items in the kit.

Suggestions

Improve delivery of kits to the clinic to avoid running low in stock. This was done during summer 2016 in two ways. First, instead of ordering kits a couple of times per year and waiting for them to be delivered, the supply of kits for a whole year was ordered at once. Secondly, vouchers were made in case the stock does run out, so that mothers can come and collect the kit later on.

Investigate the possibility to include more items in the kit, especially more plastic gloves. However this is difficult, since the kits are bought in a sealed bag. Another way to accommodate  for the lack of gloves would be to supply the clinic with extra gloves.

Make the  conditions  for receiving  a  Maama  kit  more  clear for the beneficiaries (i.e. four ANC visits at Maina clinic)

3.4 Antenatal care  attendance – the women’s perspective

One of the aims of the Maama Project is to increase ANC attendance among pregnant women. Both the CHWs and the health care staff reported an increase in the number of ANC visits and a rise in the number of women who completed four ANC visits since the project started. These reports were confirmed by an analysis of baseline  data, data from monitoring tools  and  information   collected   during   the   evaluation   interviews.

Attending four visits to receive the Maama kit

Cleaning the baby’s umbilical cord using either chlorhexidine or water and salt

Preparing baby clothes

Breastfeeding for 6 months

Preparing money for transport and emergencies

Going to the hospital or clinic for delivery or if they were not feeling well

Exercising and avoiding hard work during pregnancy

Women in the sample (n=61) who attended at least one ANC visit

100%

Baseline data from May 2013 to June 2014 obtained from the records at  Maina clinic indicated  that  on  average,  35 women  came  for an ANC visit per  month,  with 12% of women  completing  all four visits. During  the  first  project  year  (2014-2015), the  average   number   of visits increased  by  122% to  78  visits per  month,  according  to  the monthly  monitoring  tools and  cross-checked  with clinic records. The clinic records  indicated  that  82% of women  completed all four visits.

In our  study  sample  for the  year  2015-2016, 98.6% of the  women reported attending antenatal care at least once during the pregnancy, comparable to 100% in the  previous  year and  the  baseline  figure of

94% (Figure 2). A total of 85.7% of women interviewed had completed

four  antenatal  visits,  significantly  higher  than   the   baseline   figure

Project Year

Baseline

2014-2015          2015-2016

100%

98%

96%

94%

92%

    98,6%                 94%                          

90%

of 12% and  also  higher  compared to  the  estimate  from  interviews with  women  in the  previous  year,  which  was 76% (Figure 3). Most of the  women  (82.9%) attended ANC sessions  at  the  Maina Clinic.

When asked to recall five things that  they learned  from the antenatal care advice they received, most women mentioned:

Preparing themselves for birth

Eating well

Getting tested for HIV

Women in the sample (n=61) who completed 4 ANC visits

                                                                                                              100%

85,7%

76%

60%

40%

20%

12%

Sleeping under a mosquito net during pregnancy to avoid being infected with malaria

Project Year

Baseline

2014-2015          2015-2016

3.5 Antenatal care  visits –

the clinic perspective

The interviews with the nursing assistants and the clinical records gave information  about  the  content of the  ANC visits in Maina which was compared to the WHO guidelines (19).

The nursing assistants always included the following information in the clinical records: woman’s age, which ANC visit they attend, number  of pregnancies and deliveries, gestational  age, expected date  of delivery and other  diagnoses. There are no free pregnancy  tests  at the  Maina clinic; the  women  have to buy their own. An insecticide  treated  bed net is provided by the government and, if in stock, are handed out for free during the first ANC visit. Other preventive  measures  provided  by the clinic include intermittent preventive  treatment of malaria, tetanus toxoid immunization, de-worming, iron and folic acid substitution.

The only laboratory test that can be routinely performed  at the Maina clinic is the rapid Malaria test. Women are consulted  and referred to the Mayuge health  clinic for HIV and syphilis testing. There is no machine to measure  haemoglobin levels at Maina, and the only way to assess anemia  is by looking under  the  woman’s eyelids and  at  her  skin. A pregnant woman’s blood  pressure  should  be controlled  during  every visit in order  to recognize  pre-eclampsia  early, yet the  clinic’s blood pressure  machine  is currently  broken.  Screening  for malnutrition  is performed by weighing the pregnant woman during every visit. The mid upper arm circumference cannot be measured as there is no measuring tape at the Maina clinic.

The nursing assistants  give the expectant mothers  information  about pregnancy  danger  signs, breastfeeding and how to eat healthily. They also advise the pregnant women about  which items they should bring for delivery and to plan arrange transportation to the delivery facility in advance.

Challenges

It is positive that more women are coming to the Maina clinic for ANC visits and delivery, but it is challenging  for the small clinic to keep up with the increasing demand. The clinic is only open  during weekdays and at daytime, which was mentioned as a problem also by the mothers in the interviews.

Some respondents further noted  that there are too few beds in Maina, the rooms are too small and the clinic needs more health care personnel. Currently there  are not enough seats and only one antenatal/delivery bed at the clinic. Before the summer 2016 there was only one postnatal bed, but SOGH has managed to provide one more. Another challenge is the  shortage of personnel;  there  are  currently  only  two  nursing assistants, who mentioned that they would appreciate more help.

The clinic has  no  electricity  or  running  water  and  lacks important medical equipment, e.g. blood pressure machine, measuring  tape, bag and resuscitation mask. Mosquito nets and medicines (e.g. Lumartem for treating  malaria) that are provided by the government are often out of stock. Two women said during their interviews that they did not receive mosquito  nets as promised and three other women said that there was not enough free medicine. Three mothers  reported that  they did not receive the Maama kit due to it being out of stock.

One of the nursing assistants also raised the issue that the women are too young when they have children, which can lead to more complications during  pregnancy   and  delivery.  It  can  also  give  rise  to  social  and economic  problems  in the  long-term.  Five women  in the  survey said that they wanted more information about family planning.

According to the nursing assistants, the major reason  why women  do not complete  all four ANC visits is starting  the first ANC visit too late in their pregnancy. On average the women went to their first ANC visit when  they were four months  pregnant, which is later than  the WHO recommendation of 8-12 weeks.  Eleven out  of 61 women  attended their first ANC visit when  they  were 6 months  pregnant. There is no standardized plan  for what  the  ANC visits should  include  and  what information  the  mothers  should  receive. The nursing  assistants  only have the headlines in the ANC clinical records book, which is provided by the government, and their own memory to follow.

Suggestions

Larger facilities with more seating areas and at least one more antenatal/ delivery bed are needed. Electricity and running water should if possible be installed. The workforce should also be increased, preferably with a midwife. Critical items that are currently lacking should be provided to the clinic, e.g. blood pressure machine and a measuring tape. Increasing the number of Maama kits stored at the clinic would also reduce the risk of them running out of stock.

According  to  the  nursing  assistants,  it would  be  desirable  to  have an extra stock of medicine  and rapid malaria tests  in addition  to the governmental provision. The nursing  assistants  also  suggested that providing the Maina clinic with rapid HIV and syphilis tests could possibly increase  the  proportion of women  taking the  tests. The possibility to control haemoglobin levels and perform urine analyses would further add to the quality of the ANC care provided  at the clinic. The nursing assistants also suggested that the CHWs should encourage the women in the villages to go to the ANC clinic earlier during their pregnancy, so they will have time to complete all four visits.

The information given to the women about family planning by the CHWs and Maina clinic needs to be to improved. Currently, the only available family planning methods are injectables, implants and condoms.

If given resources, the  interviewed  midwife in Mayuge said that  she could  arrange  family planning  training  sessions. However, additional family planning methods should be implemented.

Additionally, the  distribution  of a mother’s card  with  the  important information about the women’s ANC visits might increase the pregnant women’s involvement. The card could also be used as a check list at the ANC visits. The nursing assistants gave positive feedback about this idea.

3.6 Deliveries

a) Delivery location and birth attendance

According to the  quantitative survey, 89% of mothers  delivered  their most recent child in a health facility, which is a marginal increase from

86% in the previous project year.  Slightly different numbers show in the monitoring data, which includes all mothers who were visited by a CHW after birth. The monitoring files state that 78% of mothers delivered at a health facility in the project year 2015-2016, compared to 67% in 2014-

2015. In any case there is a positive trend towards more facility deliveries.

The quantitative survey recorded that 46% of facility deliveries occurred in Maina clinic, where the two nursing assistants work as birth attendants,

41% in the larger health centre in Mayuge with more midwifes and 13%

in a referral hospital (see figure 5).

BIRTH LOCATION

Home                                                                          7

Health  Facility                                                              54

The choice  of facility differs from  the  evaluation  2015, where  40% delivered in Maina clinic and 60% in Mayuge health centre. The fact that more mothers delivered in hospitals might indicate a better functioning of the referral system in case of birth complications. While mothers were giving mixed accounts of the treatment in Mayuge, all respondents were very positive about the care provided by the nursing assistants in Maina. The nurses  can only offer basic obstetric  services, however  they  also provide ergometrine injections to stop  excessive bleeding  after birth. This service is also used by home-delivering mothers, as they sent family member  to get ergometrine injections from the clinic after delivery, for example when the delivery happened during the night.

Eleven percent  of mothers  delivered at home (Fig. 5), slightly less than last year (14%). Home deliveries often go hand in hand with unskilled birth attendance, delay in referrals in case of complications and unclean environments. Home deliveries can thus present a danger to the health of mother and newborn (2). The qualitative interviews revealed that while all women prefer a hospital delivery, it is not always possible. Common reasons  for home  birth were sudden  delivery onset  and thus  trouble reaching the facility. At home women were assisted by a TBA, a relative or a friend. The qualitative interviews revealed that in two cases the TBAs were a mother or a mother-in-law and were paid in cash or in kind. Both of them  had planned  to deliver at a facility, but the circumstances  did not allow it. One of the mothers used a Maama kit during home delivery.

Overall, positive trends  in delivery location  were observed. A greater proportion of deliveries occurred  in health  facilities with skilled attendance and the referral system is possibly functioning better. Even though home  deliveries have  become  less common,  a considerable portion  of women  still delivered  at  home,  often  despite  wishing  to

deliver in a facility. It is important to tackle the remaining barriers, some of which are known from similar research (20), so that every mother can deliver in the safest possible setting.

b) Birth preparedness and transportation

Data on birth preparedness and transportation to the delivery facility (Fig. 6) were newly added  to the  survey this year. 95% out  of the  42 women who used motorcycles had arranged  the transportation before the onset of labour pains. Around 80% of mothers recalled that the CHW had actively encouraged them  to save money  for transportation. The message seems to be passed on and can be seen as a project success.

TRANSPORTATION TO PLACE OF DELIVERY

Birth preparedness further includes packing a bag with essential items for delivery that  are not provided  in the  facilities. Almost all mothers brought a basin to wash the baby, baby clothes  and a jerry can with water. Almost everyone  who received  a Maama kit brought it to the delivery. Moreover, many brought clean cloths, soap, a baby blanket, tea  and  sugar. The above  indicators  were not  measured in the  2015 evaluation,  hence  no comparisons  can be drawn. The overall state  of birth- and transportation preparedness appears to be positive.

Motorcycle

78%

Walk

18%

Bicycle

4%

The transportation to the facility took on an average  32 minutes  (95% confidence  interval: 23-41 minutes). This number  is only a rough approximation,  as there were doubts  about the time estimations  given by several respondents. Moreover, 94% of women were accompanied to the delivery, most commonly by the husband, a close relative or a friend.

96% of these mothers had also previously arranged transportation back home in advance.

c) The role of traditional birth attendants (TBAs) in delivery

This year we were also able to interview a traditional  birth attendant (TBA) in the village of Bulondo to find out more about the role of TBAs in maternal health care, particularly around home births. Since 2010, TBAs have officially been banned in Uganda, though they continue to practice given the poor implementation of the ban (27). The TBA perceived that since the community  knows of and trusts them,  they are still seen  as relevant people in the community, though many TBAs increasingly refer women to health care centres for delivery.

The TBA further explained  that  she had been  in this role for about  15 years; she  had  only recently  moved  to Bulondo, but  in her previous location  she saw on average  100 mothers  in a month.  The common practice was that the expectant mothers  would come to her (unless it was an emergency, in which case she would go to their houses) and this happened most often in situations where they could not access care at a health  facility, due to time or geographical constraints. Items for the delivery (gloves, razor blade, sheet, thread  for cord, basin, soap, sugar, pads, clothing for baby, diapers) as well as food are usually brought by the mother; in some cases the TBA would provide food as well.

Before the delivery, the TBA would initially assess the mother’s condition to determine if the birth could take place at home, or if a referral was needed to  the  nearest  health  facility.   If  the  TBA  could  handle  the situation, she would assess the approximate time when to expect  the baby, and make the mother  feel comfortable  and provide tea and food until the time of delivery. The TBA would then  assist with the delivery and monitor the health  of the mother  and baby for the following four hours. If the delivery takes place at night, the mother usually stays over and leaves in the morning. In terms of post natal care, the TBA follows up with check ups on the mother  one day and again one week after birth and also gives advice on family planning.

Challenges

Expecting mothers usually know about and plan to deliver at a health facility. However, practical problems such as transport to the health facility can not be overcome in the current project and lead to that some mothers decide to deliver with a TBA.

Limited capacity of Maina clinic. At the start of the evaluation there was one bed for delivery and one bed for resting afterwards. This meant that if two women had to rest after delivery, one had to rest on a mat on the ground. both mothers and health care staff brought this up.

Recommendations

In the long term, it should be considered if SOGH can organise a system to make it more easy for mothers to get to the clinic. For example, a bodaboda that mothers in need can call.

The limited capacity of Maina clinic was remedied during the evaluation by the purchase of an extra bed where women who have delivered can take rest before making the journey home.

3.7 Newborn  health

Almost half of the newborns  (44%) had no health problems during the neonatal  period of 28 days. The reported problems  for the rest of the newborns  included malaria (36%), colds and coughs (11%), skin rashes (8%), problems with the umbilical cord (7%) or a combination of these.

a) Newborn care practices

We recorded  the  prevalence  of several good  newborn  care practices which are recommended by WHO (25). 92% of mothers  had the baby put on the bare skin of their chest directly after delivery. Close to  all mothers  breastfed  their children. Ninety-three  percent  started  within one hour from birth, which is more than in the 2015 evaluation  (59%). Exclusive breastfeeding in the  first three  days of life was reported by

83% of mothers, compared to 79% in 2015. The ten mothers  who gave their babies something else to drink than breastmilk used warm water, sometimes with sugar. Four women did not produce enough breastmilk and thus needed to supplement. All mothers  except for two were still breastfeeding when being interviewed, when over 30% of the children were older than 6 months. Long-lasting breastfeeding appears to be the norm in the project area and may contribute to birth spacing.

Overall, direct skin contact with the baby and immediate, exclusive and long-term  breastfeeding appear  to be strong  social norms within the surveyed  population.  This was seemingly  already the  case before  the project start in 2014. Yet, this evaluation  revealed a further increase in immediate and exclusive breastfeeding compared to 2015. Emphasizing the importance of good care practices during CHW home visits and ANC has likely contributed to this positive development. In contrast, other reports on newborn care in Uganda suggest a more problematic situation (8). The next evaluation could benefit from measuring  a wider range of safe newborn  practices to identify the areas that  need  improvements also in the project area.

b) Umbilical cord care

Of the  mothers  in the  project  area, 36% received  the  umbilical cord disinfectant  chlorhexidine at their fourth antenatal visit. The low proportion could be explained  by that  the chlorhexidine  intervention was initiated in March 2016 and by then many of the mothers  already had their fourth ANC visit or their delivery.

Out of the  women  who  received  chlorhexidine, 96% (21 out  of 22) applied it. The mother who did not apply it explained that she forgot to use it. Out of the women who received and applied chlorhexidine, 81% (17 out of 21) applied it within 24 hours. An important note is that many mothers explained that they applied chlorhexidine multiple times, while they were given a tube  for a single time application. This could thus mean that the chlorhexidine is not optimally used. Mothers who did not receive chlorhexidine reported no specific method of umbilical care, or that they washed the stump  with water or warm water, in many cases with added soap or salt.

Due to a small sample study, no conclusions can be made if the newborns receiving  chlorhexidine  experienced fewer  infections. However, it is known that chlorhexidine effectively decreases umbilical cord infections in low-income settings (26). The finding of one infection and three slowly healing cords in our small sample suggests  a need for intervention. The qualitative  interviews  revealed  that  mothers  had  a positive  attitude towards  chlorhexidine  and said they would like to use it, if available. Thus, SOGH currently discusses how chlorhexidine  provision might be permanently incorporated into the Maama project. Ensuring the gel’s correct application will be a further challenge.

Challenges

Stock of chlorhexidine: At the time of the evaluation the Maina clinic had no stock of chlorhexidine. Because of a miscommunication with the field project manager, the rest of the stock had not been delivered to Maina clinic.

Six of  the  women  who  reported  having  received  chlorhexidine had delivered before the intervention was started. Two of them delivered at Mayuge health center, where it may have been possible that they received chlorhexidine. The other four delivered at Maina, where we have  no knowledge of available chlorhexidine at that time. These answers might be due to a social desirability answer in some cases, in other cases these women  talked about when and how they applied it, so there possible explanation is that these women are not always fully informed about what they receive.

Use of chlorhexidine: The chlorhexidine is supposed to be used as a one time  application. However, some  of  the  mothers  reported that  they used it multiple times during multiple days. Multiple use of the received chlorhexidine makes it doubtful that a sufficient amount is used. One of the mothers used the chlorhexidine three times a day for one week, but reported that her child’s cord healed slowly,which could be an indication of unsatisfying results resulting from misuse of chlorhexidine. The misuse could in some cases be explained by that the pictorial instruction was not given to all mothers.

Recommendations

The CHW have received extra training to be explain the use and application of chlorhexidine to ensure better use in the future. The Maina clinic also has a pictorial on the wall explaining how it should be used.

3.8 Male involvement

In order to improve the results and to sustain the Maama Project in the long run, both mothers and fathers need to be involved. An exploratory investigation was conducted by interviewing four husbands to mothers who took part in the Maama project. The respondents were accessible during daytime, which may contribute to selection bias since the men who stay at home for work, in comparison to men who work away from home, are likely to be more positive to the project due to a higher degree of information and involvement.

As the  communities  have  become  sensitized  to  the  Maama  Project, CHWs have reported that  men  are increasingly positive towards  ANC visits and  the  work of the  CHWs. One important reason  behind  the positive attitude is that  the project and the Maama kits not only help the women, their unborn and newborn babies, but also decrease men’s work that otherwise would include retrieving items for the birth which is a costly and time-consuming venture.

With sensitization, the men stated  that they are more concerned about women’s and  newborn’s health  and  they  expressed  support  of the CHWs, ANC and the project  overall. One stated  reason  for supporting the project was that several fathers now feel less worried when working out of the home as they know that the CHW will be there supporting the wives.

All male respondents stated that they participated in birth preparedness with a CHW and/or  at ANC visits at a clinic and could elaborate  on and specify the ways in which they and their families had benefited from the visits. Examples of new insights included additional ways to support their wife during pregnancy  and after giving birth, encouragement of mothers to attend ANC and to give birth at a health facility and additional knowledge about child care. Three out of four fathers had attended ANC

visits with the mother. Of the mothers  participating  in the survey, 41%

reported that their husband had accompanied them to the delivery.

Challenges

Potential  challenges  for male  involvement  include  strong  norms  of household responsibilities and family roles within the communities. One challenge  brought up in some CHW interviews is that men sometimes have  had  a bad  attitude when  the  CHW goes  to see  the  mother  or that men encourage their wife to go to a TBA. Lack of information and understanding of the purpose of the project was suggested to contribute to the men’s behaviour. In addition,  there  have been  occasions when the mother  hides the Maama kit from the husband in order to sell the items or to get additional money from her husband (to buy items for the child birth) although she already has gotten a Maama kit for free from Maina clinic. Traditional power  structures  and lack of communication between the parents  and the CHW can thus be a cause of disruption within families. CHWs have also reported that some men need  to take more  responsibility  during  the  pregnancy  period  by obtaining  items such as clothes for the baby as well as providing transportation money for the mothers to the health facilities.

Suggestions

Recommendations forward for the Maama Project include continuous spread  of information  in the communities, primarily in the rural areas, with an emphasis to address the importance of men in understanding and  taking full responsibility  as fathers  and  not  perceiving  initiatives such as the  Maama project  as a way to avoid responsibility. It is also important  to  discuss  how  the  family  and  the  community   overall benefit from increased male involvement. An identified opportunity to strengthen male involvement is to increase incentives for the fathers to be more engaged in their wife’s pregnancy. Finally, it is of importance to strengthen and support the group of single mothers.

3.9 Mental health

Research conducted in Uganda  indicates  the  existence  of maternal mental  health  issues  and  postpartum depression (21).   In order  to investigate whether mental health was a field of interest for the Maama Project we created  a qualitative tool aimed at CHWs as well as adding questions  in the  quantitative tool  aimed  at  the  mothers. Both tools were loosely based  on the Edinburgh scale. After trial in the field and consulting  the  UDHA staff the  tools were modified  towards  focusing mainly on  behaviours  as symptoms  of mental  health  problems. Our findings indicate that 16% (10 women) reported that they experienced one or several signs of maternal mental health problems or postpartum depression.

However, out of these women 50% reported that this was due to physical problems such as malaria or pains after the delivery. Of the women who had experienced at least one sign of postpartum depression 70% (7 out of 10 women) did not desire more children which is considerably higher than, 35% among  the women  who had not experienced any signs of postpartum depression.

Only one of the CHWs said that  she had met a woman  suffering from postpartum depression,  one  other  CHW said that  the  women  would talk to her if they were sad after a miscarriage or a stillbirth. This result should be interpreted with precaution;  several factors could have had an effect on the validity of the study, such as stigma related to mental health problems (21) and the tool not being culturally sensitive enough, causing a failure in screening for maternal mental health problems.

3.10 Family planning

According to WHO, the  promotion of family planning, and  ensuring access to preferred  contraceptive methods for women  and couples, is essential to securing the well-being and autonomy of women. Further, WHO claims that the use of family-planning can improve both maternal and infant health as it can prevent closely spaced, high-risk and ill-timed pregnancies and births (22).

According to the quantitative study carried out, 23% (14 women) of the women interviewed reported that they were currently using a modern contraceptive method (breastfeeding not included). Out of these women the most commonly used contraceptive methods were injectables (8.2

%) or condoms  (6.6%). Out of the women who did not currently use a family planning  method, 52% (24 women) clearly expressed  that  they intended to use a family planning method later on.

This result is in line with the findings from the qualitative  interviews with the  fathers  who,  despite  of desiring  more  children, either  used or intended to use a contraceptive method in order to delay the next pregnancy. Among the  women  who intended to use a contraceptive method later, the most common methods mentioned were injectables, male condoms or female sterilisation.

Out  of  the  women  who  were  currently  not  using  a  contraceptive method 10% (5 women) expressed that they were not planning to use a contraceptive-method at all.

Many women replied that they did not use a contraceptive method as it was too soon after the delivery; they either expressed that they were using postpartum abstinence, that they needed to heal, that they were breastfeeding or that they were waiting for their period to come back.

Out of the women who intended to use a family planning method later on, five women  expressed  that  they had concerns  regarding  the side effects of certain family planning  methods or that  they needed more information. This made  them  delay their use of or reluctant  to find a method that they thought suited them.

Half of the women (31 women) in our survey reported that the CHW had talked to them  about  family planning. Most women (16) were advised to use family planning  or to go to the clinic/hospital for contraceptive counselling in order to find a suitable contraceptive method. Furthermore, most of the women were also given general information about family planning methods and that they could be used in order to avoid getting pregnant.

Challenges

The findings  clearly indicates  that  most  women  were  aware  of the existence  of different  family planning  methods. Using the  data  we acquired from our survey it is hard to know whether  the women were able to space or limit their fertility in accordance with their own wishes. A suggestion is to further investigate the existence of unmet contraceptive needs among the beneficians of the Maama Project. Education, receiving information about the contraceptive method of choice, communication with one’s partner, and finding a suiting contraceptive method has been proven to lead to a continuous contraceptive use [23,24].

Recommendations

More knowledge  about  family planning  methods was requested by some of the CHW. Continuous sensitizing of the community about family planning  methods, including  the  men,  and  possibly  the  CHWs,  can therefore  be suggested. As certain women experienced negative  side- effects from their previous contraceptive method (notably injectables), it is recommended that other contraceptive methods, for example non- hormonal contraceptives, should be made available at Maina clinic.

The Maama-project evaluation team 2016

30

References

1. World Health Organization, World Bank Group, United Nations (2015). Levels & trends in child mortality 2015. New York: UNICEF.

2. Lawn  JE, Cousens S, Zupan J. 4 million neonatal deaths: When? Where? Why? Lancet. 2005 Mar 11;365(9462):891–900.

3. Lawn JE, Kinney MV, Black RE, Pitt C, Cousens S, Kerber K, et al. Newborn survival: a multi-country  analysis of a decade  of change. Health Policy Plan. 2012;27(suppl 3):iii6–28.

4. Lawn JE, Blencowe H, Oza S, You D, Lee AC, Waiswa P, et al. Every

Newborn: progress,  priorities,  and  potential  beyond  survival. Lancet.

2014 Jul 18;384(9938):189–205.

5. UNICEF, World Health Organization. A decade of tracking progress for maternal, newborn  and child survival. The 2015 report. Geneva: World Health Organization; 2015.

6. Bhutta ZA, Das JK, Bahl R, Lawn JE, Salam RA, Paul VK,  et al. Can available interventions end  preventable deaths  in mothers, newborn babies, and stillbirths, and at what cost? Lancet. 2014;384(9940):347–70.

7. Waiswa P, Kallander K, Peterson  S, Tomson G, Pariyo GW. Using the three  delays model to understand why newborn  babies die in eastern Uganda. Trop Med Int Health TM IH. 2010 Aug;15(8):964–72.

8. Waiswa P,  Peterson  S, Tomson G, Pariyo GW.  Poor  newborn  care practices-a population based survey in eastern Uganda. BMC Pregnancy Childbirth. 2010; 10(1):9.

9. WHO, UNICEF. Home visits for the newborn child: a strategy to improve newborn survival. 2009.

10. Kirkwood BR, Manu A, ten Asbroek AH, Soremekun  S, Weobong  B, Gyan T, et al. Effect of the Newhints home-visits intervention on neonatal mortality rate and care practices in Ghana: a cluster randomised controlled trial. The Lancet. 2013 Jun;381(9884):2184–92.

11. Hanson C, Manzi F, Mkumbo E, Shirima K, Penfold S, Hill Z, et al.

Effectiveness of a Home-Based Counselling Strategy on Neonatal Care and Survival: A Cluster-Randomised Trial in Six Districts of Rural Southern Tanzania. PLoS Med. 2015 Sep 29;12(9):e1001881.

12. Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based neonatal  care  and  management of sepsis  on  neonatal mortality: field trial in rural India. The Lancet. 1999 Dec;354(9194):1955–

61.

13. Bang AT, Reddy HM, Deshmukh MD, Baitule SB, Bang RA. Neonatal and Infant Mortality in the Ten Years (1993 to 2003) of the Gadchiroli Field Trial: Effect of Home-Based Neonatal Care. J Perinatol. 2005;25(S1):S92–107.

14. Kumar V, Mohanty S, Kumar A, Misra RP, Santosham  M, Awasthi S, et al. Effect of community-based behaviour  change  management on neonatal mortality in Shivgarh, Uttar Pradesh, India: a cluster-randomised controlled trial. The Lancet. 2008 Sep;372(9644):1151–62.

15. Baqui AH, El-Arifeen S, Darmstadt GL, Ahmed S, Williams EK, Seraji HR, et al. Effect of community-based newborn-care intervention  package implemented through  two service-delivery strategies  in Sylhet district, Bangladesh:  a  cluster-randomised  controlled  trial. The  Lancet.  2008

Jun;371(9628):1936–44.

16.  Bhutta ZA, Memon  ZA, Soofi S, Salat MS, Cousens  S, Martines J. Implementing community-based perinatal  care: results  from  a  pilot study in rural Pakistan. Bull World Health Organ. 2008 Jun;86(6):452–9.

17. Waiswa P, Pariyo G, Kallander K, Akuze J, Namazzi G, Ekirapa-Kiracho E, et al. Effect of the Uganda Newborn Study on care-seeking and care practices: a cluster-randomised controlled trial. Glob Health Action. 2015

Mar 31;8:5–15.

18. Waiswa P, Namazzi G, Kerber K, Peterson  S. Designing  for action: adapting and implementing a community-based newborn care package to  affect national  change  in Uganda.  Glob Health  Action. 2015 Mar

31;8(0).

19. Lincetto O, Mothebesoane-Anoh S, Gomez P, Munjanja S. WHO 2006, Opportunities for Africa’s newborn, chapter 2 Antenatal care.

20. Mrisho M, Schellenberg  JA, Mushi AK, Obrist B, Mshinda H, Tanner M, Schellenberg  D. Factors affecting  home  delivery in rural Tanzania. Tropical Medicine & International Health. 2007 Jul 1;12(7):862-72.

21. Baron E, Hanlon C, Mall S, Honikman S, Breuer E, Kathree T, Luitel N, Nakku J, Lund C, Medhin G, Patel V, Petersen I, Shrivastava S, Tomlinson M. Maternal mental health in primary care in

five low- and  middle-income   countries:  a  situational  analysis.  BMC Health Services Research. 2016; 16:53.

22. World Health Organization (2015). Family Planning/  Contraception Fact  Sheet  N°351. Retrieved  from: http://www.who.int/mediacentre/ factsheets/fs351/en [06/12/2016]

23. Sanogo, D., RamaRao S. , Jones H., N’diaye P., M’bow B., Bamba Diop C. (2003). Improving Quality of Care and Use of Contraceptives in Senegal. Source: African Journal of Reproductive Health / La Revue Africaine de la Santé Reproductive. 2003 Aug; 7(2):57-73.

24. Sileo, K.M., Wanyenze, R.K., Lule, H. , Kiene, S.M. (2015). Determinants of family planning  service uptake  and  use  of contraceptives among postpartum women in rural Uganda. Int J Public Health (2015) 60:987–

997.

25. World Health Organisation, 2010. Essential Newborn case Practice, Clinical Practice guide. Geneva. Retrieved from: http://apps.who.int/iris/ bitstream/10665/70540/4/WHO_MPS_10.1_Clinical_practice_eng.pdf

26. World Health Organisation, Second Meeting of the Subcommittee of the Expert Committee on the Selection and Use of Essential Medicines Geneva, 2008. Review of the  available evidence  on 4% chlorhexidine solution  for umbilical cord care. Retrieved from: http://www.who.int/ selection_medicines/committees/subcommittee/2/chlorhexidine.pdf

27. Alanna Shaikh, 2010. UNIDISPATCH. Traditional birth attendants  in

Uganda. Retrieved from: http://www.undispatch.com/traditional-birth-

attendants-persist-in-uganda/                                                                          Pictures taken by the Maama-project evaluation team 2016

32