Family Planning Behaviors, Preferences, Needs and Barriers Within Priority Geographies

Consumer-focused Research to Accelerate Voluntary Uptake of Modern Contraceptives: Using Novel Data and Artificial Intelligence to Assess Consumer Intent, Identify Barriers and Enablers, and Uncover High-Impact Practices in Sindh, Pakistan

This is a multi-country cross-sectional study that will use mixed methods to better understand the extent to which women of reproductive age have intention to use modern contraception, what barriers and enablers exist, and what practices would support them to voluntarily adopt a modern contraceptive method. The study will take place in four countries i.e., Senegal, Côte d'Ivoire, Nigeria, and Pakistan, each with an adapted quantitative tool that will be administered to women of reproductive age and to men who live in their households. Qualitative data collection will be used to better understand the results of the quantitative survey.

The study aims to identify the proportion of women who have the intention to use family planning but are not currently using modern contraceptive methods and to determine why women with FP intent are not adopting modern contraceptive methods. Further it will also assess the characteristics of women who show intention-action gaps, and the interventions that enable them to adopt a modern contraceptive method. The findings of this study will support on-the-ground programs in identifying which population segment a woman might belong to.

The research will address its objectives by addressing the following research questions:

1. Who has the intention to use family planning but is not currently using modern contraceptive methods?
2. Why are women with FP intent not adopting modern contraceptive methods? Which methods are preferred, accessed, and intended?
3. Who should be targeted to diminish intention-action gaps, and how?
4. How can on-the-ground programs identify which population segment a woman might belong to?

2 INTRODUCTION It has been estimated that for every $1 invested in family planning, society reaps $120 in lifetime benefits spanning health, economic and social advantages. Modern contraception can prevent unintended pregnancies and reduce the number of unsafe abortions, maternal mortality, and health problems arising from complications in pregnancy and childbirth. Modern contraceptives help couples effectively space childbirth and assume greater control over their preparedness for parenthood. Increased contraceptive uptake has been associated with social benefits such as higher education attainment among girls, women's prolonged participation in economic activities and increased agency in household decision-making. The ripple-on impacts include poverty reduction and improvement in the GDP growth rate. At a national and global level, modern contraception is vital for achieving the Sustainable Development Goals.

Despite the advantages of effective modern contraceptives and investments in the past decade, uptake remains slow in Sub-Saharan Africa (29%) and South Asia (52%). These regions face both higher maternal and infant mortality rates and lower female literacy rates along with some of the highest poverty rates in the world. Modern contraceptive uptake varies across countries and within each region, with some seeing concentrated vulnerability. Modern contraceptive prevalence rate (MCPR) in Senegal and Ethiopia is 26% and 25% respectively) and even lower in Côte d' Ivoire and Nigeria (18% and 12% respectively). The adoption of modern contraceptive for spacing is lowest among younger women who have a greater need for modern contraceptives. Similarly in South Asia, Pakistan (25%) has a lower MCPR than other countries in the region such as Bangladesh (59%) and India (56%). Within these areas exist significant populations of women who want to use modern contraceptives but are constrained by a range of barriers that require greater understanding and precise targeting.

To maximize the impact of dollars invested in family planning, focusing interventions on women who have family planning (FP) uptake intent, and targeting them with "high impact practices" - evidence-backed FP uptake drivers - will likely have some of the biggest and swiftest returns, and may yield shifts in broader norms. Recent studies have shown that intention to use within a specified period (e.g., 12 months), is a far better predictor of modern contraceptive use than either unmet need, often a core outcome metric, or even intention to use without a specified timeframe. However, there is a lack of foundational data to deeply measure intent, understand what drives intent, and how that varies for different groups (aka segments) of the population. This foundational data is needed to identify populations and geographies that may yield swift and significant improvements in modern contraceptive uptake, the exact barriers to address, and appropriate channels to address different groups of women.

2.2 STUDY RATIONALE This study will take a consumer-centered approach to understanding women's intention to use modern contraception methods and the barriers they face in using them. This study will generate new data that will allow government and their local partners to improve planning for programs that support women in voluntarily taking up modern contraceptive methods, leading to an increase in FP uptake in the near term. To generate this data, the investigators will undertake a multi-step process.

1. Collect comprehensive formative data on drivers of FP intention and uptake. To determine priority areas for intervention, the investigators will collect novel data that holistically capture behavioural and contextual barriers to the uptake of modern contraceptives. The study will gather a 360-degree view of potential drivers of women's FP decisions in each geography from key target populations, focusing on adult women of reproductive age. Using Surgo's peer-reviewed CUBES frameworks and years of family planning survey experience, the investigators will use innovative methods to capture key barriers and preferences across contexts through in-depth quantitative survey. These surveys will also include an innovative modular index aiming to add precision to the measurement of FP intent. These questions will be iterated through cognitive testing through qualitative interviews utilizing the quantitative questionnaire, and will be validated using data collected in the first phase of the project.
2. Apply advanced AI-powered analytics to drive strategy. (1) uncover the who, why, where, and how of underlying consumer forces that drive modern contraceptive intent and use will be uncovered, and (2) identify distinct segments with varied needs and barriers within the target population.. Based on the sample size and analytical approach applied, insights will be disaggregated at different geographical levels (described in the Sampling section of this document) to aid meaningful programmatic targeting on the ground.
3. Co-develop analytics-driven solutions with implementation partners. The precision of the survey findings will allow AKU with sponsor, in collaboration with government partners and local partner, to generate segment-specific intervention recommendations that draw from, refine or add to high-impact practices identified by the FP High Impact Practice (HIP) consortium. This will include generating data-driven inputs for on-ground teams to design and strategize service approaches and target the right HIP to the right individuals. Findings will be shared with key government and partner stakeholders to support implementation at scale via a variety of mechanisms. AKU and Surgo will also support innovation uptake through concurrent learning and iteration with program partners and government stakeholders to guide insight use.

3. STUDY OBJECTIVES

The research has three primary objectives:

1. To identify the proportion of women who have the intention to use family planning but are not currently using modern contraceptive methods.

   Endpoint: Predictors of modern contraceptive intent and uptake in women meeting the inclusion criteria

2. To determine why women with FP intent are not adopting modern contraceptive methods Endpoint: Context-specific barriers at individual and structural levels, data-driven patterns of causation, and a comparison of the impact strengths of high-impact practice on various determinants.

3. To identify the characteristics of women who show intention-action gaps, and the interventions that enable them to adopt a modern contraceptive method Endpoint: ML-derived distinct population segments that differ in their preferences and attitudes towards modern contraceptive methods, relative prevalence of different segments and high-impact interventions matched to each segment.

The secondary objective of the research is to support on-the-ground programs in identifying which population segment a woman might belong to.

Endpoint: Typing tool with simple questions that can be deployed via C/LHWs and other field interventions 4 STUDY DESIGN This will be a cross-sectional study, employing a mixed-methods approach, comprising a representative quantitative survey of women of reproductive age. A complementary qualitative data collection that will be carried out after the quantitative survey. The post-baseline qualitative interviews will be conducted with a small subset of women who participated in the quantitative surveys, and will allow a deeper and more nuanced understanding of the quantitative findings.

Quantitative data will be collected via a comprehensive and contextualized survey, with modules designed to:

• Understand a woman's demographic characteristics
• Precisely measure intention around FP use among varied women at different life stages and in different contexts
• Identify drivers of intent-action gaps, to better support women with intent to use contraceptives
• Highlight pathways for demand generation and uptake for FP among those without current FP intent
• Gather data on product and method preferences and satisfaction to guide deployment, introduction, and continuation support efforts Baseline survey data collection will be followed by focused, rapid qualitative interviews with a subset of respondents. Specific areas of inquiry will be identified post-survey data analysis, as these interviews will be aimed at disentangling unexpected survey findings.

4.1 SAMPLE SIZE: A divisional-specific survey will be conducted in Sindh province. To calculate the sample size for this study on modern contraceptive use among married women of reproductive age (MWRA), the calculations are based on several key assumptions: a design effect of 1.8202 to account for survey complexity, a relative margin of error (RME) of 0.15 for precision, and the target population constituting 15.3% of the total population. With an average household size of 6.1 and an individual and household response rate of 0.9, the number of households per primary sampling unit (PSU) are determined to be 12. Using the percentage use of modern contraceptives and the correlation coefficient (r) for each division, the required number of individuals, households, and PSUs were calculated to ensure a representative sample, resulting in a total of 9230 individuals, 10970 households, and 914 PSUs. See table Assumptions Target population: MWRA Design effect: 1.8202 Relative Margin of Error (RME): 0.15 Size of target population from total population: 15.3% Average household size: 6.1 Individual and HH response rate: 0.9 HH per PSU: 12 Sample size Division % of use of modern contraceptive R Individual HH PSU Hyderabad 23.6 0.236 1164 1383 115 Karachi 24.4 0.244 1114 1324 110 Larkana 17.3 0.173 1719 2043 170 Mirpur Khas 12.5 0.125 2517 2991 249 Shaheed Benazir Abad 19.8 0.198 1456 1731 144 Sukkur 22.2 0.222 1260 1498 125 Total 9230 10970 914

Definitions:

Primary sampling unit: A PSU will be determined using GIS technology. This will consist of around 1000 population.

Household: A household consists of one or more persons who live in the same dwelling and share the same kitchen.

4.2 SAMPLING STRATEGY

In Sindh, which has a total of six divisions, and 30 districts. A multistage random sampling approach was employed. The following steps were adopted:

• Divisional level sample size estimates. Estimates of sample size to ensure results are representative at both the divisional and provincial levels in Sindh province, using the aforementioned assumptions.
• Distribution of divisional level estimated sample size among districts of the respective Divisions: the sample size will be distributed among districts within respective regions using a probability proportional sampling (PPS) approach.
• Selection of PSUs within districts: Employ GIS technology to create clusters and perform random selection of PSUs.

Select Households within PSUs: Use a systematic sampling approach to select 12 households within each PSU using fresh household line listing data In each household, enumerators will begin the interview by seeking verbal permission from the household head (or another knowledgeable adult household member), and will create a roster of household members. One married woman of reproductive age in the list of household members will be randomly selected for interview. The written consent/assent will be administrated with that selected woman for a detailed interview.

4.3 SURVEY INSTRUMENTS Three main survey tools will be designed: the main quantitative questionnaire, the questionnaire for male respondents, and the qualitative interview guide. The quantitative tools will be programmed and administered through computer-assisted personal interviews (CAPI).

4.3.1 Quantitative tools The quantitative survey instruments will cover a range of factors proven to influence intent and FP behaviors, summarized below.

Theme Sub-theme Intent to use contraceptives Intent intensity, time horizon, milestone linkage, and perceived implementability Perceptual or behavioral drivers Awareness, beliefs and perceptions, preferences, biases, heuristics, emotions, self-efficacy and intention around modern contraceptive uptake Contextual or system-based drivers Access, cost, health system interaction, social norms, incentives and socio-demographics Influencers Social networks, family decision-making, information channels and media usage Consumer perceptions of delivery of family planning service by the healthcare system FP visits by CHWs, inability to access contraceptives due to stock-out or non-availability of providers Experience with the healthcare system Utilization of services, perceptions of the quality of counseling and care offered by CHWs and healthcare facilities Product and method preferences Product-level perceptions, misconceptions, peer perspectives, preference drivers, and usage and continuation/discontinuation history Given that family planning behaviors are contextually specific and differ widely between groups within and between geographies, the intend is to use pre-survey qualitative cognitive interviews to guide design and questionnaire contextualization in each country, recruiting individuals to represent varied internal contexts. Key aspects investigated will include social, gender, FP availability, media exposure, and contraception perceptions. The surveys will be standardized to the degree possible, with limited variability based on country factors. The cognitive interviews and qualitative work will inform not just survey content but phrasing and translation.

A gender-responsive lens will also be used across the questionnaire design and fieldwork. In the survey, it will specifically focus on women's participation in the decision to use modern contraceptives. By exploring factors such as autonomy, domestic violence, gender norms, spousal support, access and mobility, media exposure, socioeconomic factors, parity and cultural and social norms, the diverse dynamics that influence women's self-efficacy will also be explored.

To ensure inclusivity and comfort during fieldwork, the research lead will pre-test survey comprehension and engage female enumerators for conducting interviews at the respondents' homes or other private locations. The investigators have extensive experience conducting data collection on sensitive topics from family planning to abortion and will include gender sensitivity training for all enumerators. Also take specific precautions will be taken to ensure utmost sensitivity to the respondent and moderators' safety, privacy, and comfort. To triangulate survey responses, surveys with a small sub-sample of husbands or mothers-in-law will also be conducted.

4.3.2 Post-Baseline Qualitative Interviews Baseline survey data collection will be followed by focused qualitative interviews with a subset of respondents. Specific areas of inquiry will be identified post-survey data analysis, as these interviews will be aimed at understanding the underlying factors behind certain behaviors for a nuanced understanding of the complexity of the reported phenomenon.

Prior to qualitative data collection, semi-structured interview guides will be developed to address these areas of inquiry, which will be used to guide conversations and ensure a comprehensive exploration of relevant topics. This will be submitted with ERC as an amendment and used in the field after approval.

5 STUDY POPULATION 5.1 INCLUSION CRITERIA All married women aged 15-49 living in the sampled PSU and providing written informed consent are eligible to participate in the study. Only one MWRA will be selected from one household.

Inclusion criteria for qualitative interviews will depend on the specific areas of inquiry determined by the analysis of baseline survey data; however, only women who have participated in the baseline survey will be eligible to participate in the post-baseline qualitative interviews.

5.2 EXCLUSION CRITERIA

• Women below 15 years and above 49 years are not eligible to participate
• Women aged 15-18 will be excluded unless a parent, guardian, or spouse consents to her participation in the study,
• Participants who do not speak the languages in which the surveys and interviews are conducted may be excluded to ensure accurate communication and understanding of the study materials.
• Individuals with severe cognitive impairments or intellectual disabilities that could affect their ability to comprehend the study objectives or provide meaningful responses may be excluded.
• Participants who are unable or unwilling to provide informed consent for participation in the study will be excluded to uphold ethical standards and protect participant rights.

5.3 STRATEGIES FOR RECRUITMENT AND RETENTION Enumerators will undergo comprehensive training on approaching potential participants with sensitivity and respect, clearly describing the study objectives, procedures, and potential benefits to encourage participation. They will be equipped with effective communication skills to address any concerns or questions participants may have and emphasize the importance of their contribution to the research. Additionally, enumerators will be instructed on strategies to encourage completion of the survey while ensuring participants feel empowered and comfortable throughout the process. This approach will foster trust between enumerators and participants, ultimately contributing to successful recruitment and retention efforts. Importantly, enumerators will be trained to acknowledge and respect participants' rights to refuse participation or withdraw from the study at any stage without repercussion. Respondents will be given a small non-monetary incentive in return for the time they spend participating in the interview. This may be a small amount of phone credit or another household item. The value of this item is expected to be approximately PKR 500-1000. This will be given after the completion of the interview, even if the participant decides not to complete the questionnaire in full.

5.4 STUDY DISCONTINUATION/WITHDRAWAL Participation in the study is entirely voluntary, and participants retain the right to withdraw at any stage without consequence. Should participants wish to withdraw from the study during the survey, they can simply inform the enumerator, and their data up to that point can be discarded. However, due to the anonymous nature of survey responses, data cannot be deleted after completion.

After baseline data collection, participants selected for interviews can choose whether or not to take part in the interviews, and they can terminate their involvement at any time by communicating their decision to the interviewer.

6 STUDY PROCEDURES 6.1 STUDY SITES This study will be carried out in all districts of Sindh. The selected geographies will be determined later as described in the sampling methods.

6.2 PARTICIPANT RECRUITMENT 6.2.1 Household listing The research team will allocate randomly selected Primary Sampling Units (PSUs) to the field staff. These PSUs will serve as the focal points for the data collection process. The field team will then carry out household listings within these PSUs, meticulously documenting information about each household. This comprehensive approach aims to identify and enumerate potential participants for the survey, ensuring a representative sample from the target population.

6.2.2 Baseline surveys For the quantitative baseline survey, participants will be recruited by a field team member who will visit their household in person when a household is selected according to the sampling approach described in 4.2. Recruitment will be done after administration of written informed consent or assent.

6.2.3 Post-baseline qualitative interviews During the quantitative baseline survey interview, respondents who complete the interview will be asked for their permission to retain their contact information and to be contacted for a possible follow-up interview. A database of the names and contact information of those who consent will be created.

During the data analysis, selection criteria for the qualitative interviews will be created based on areas of inquiry that arise based on the data collected in the quantitative data collection. The database of consenting women will be used to purposively identify potential respondents for the qualitative interviews. Wherever possible, potential respondents will be called by phone to set up an interview time that will be convenient for them, though this may not be possible in areas where rates of phone ownership are low or where cellular networks are unreliable. Respondents to the qualitative interviews will be eligible to receive additional compensation for their time.

6.3 DATA COLLECTION 6.3.1 Baseline surveys Quantitative interviews will be conducted via face-to-face interviews using a structured electronic questionnaire (CAPI). Before data collection begins, participants will be briefed on the purpose of the study and verbal/written informed consent will be obtained.

Interviews will take a maximum of 45-60 minutes to complete. Participation will be voluntary and participants can end surveys at any time; however, enumerators will be asked to encourage survey completion. All questions except branch logic questions will be mandatory, but respondents can choose "prefer not to answer" options for areas they prefer to keep confidential. The majority of questions will be multiple choice, multiple answer and rating scale questions, though strategic use of open-text responses will be undertaken.

6.3.2 Post-baseline qualitative interviews Post-baseline qualitative data collection will be conducted through semi-structured interviews covering themes based on preliminary analysis of the quantitative baseline. Participants will be briefed on the purpose of the study and their consent will be obtained prior to beginning the interview. Each interview will be audio-recorded with the participant's permission to ensure accurate data capture. Interviewers will follow the guide but also allow for flexibility to explore emerging themes and topics raised by the participants. (Interview guides and consent form for these qualitative interviews will be submitted later when the quantitative data will be analyzed and themes will be generated for qualitative assessment. These cannot be generated at this point) The environment will be set up to ensure privacy and comfort, encouraging open and honest communication. Field notes may be taken to capture non-verbal cues and contextual information. Following the interviews, recordings will be transcribed verbatim and translated into English for subsequent analysis.

6.4 QUALITY ASSURANCE AND QUALITY CONTROL Data quality assurance and quality control will be an ongoing process throughout the field work and analysis. Data quality control processes will be built into the programmed questionnaires and incorporated into the training of the data collection agents.

To ensure data quality before the analyses, investigators will take rigorous steps to prevent respondent bias and aim to avoid respondent fatigue while maximizing information. Cognitive interviews will be conducted to ensure appropriate question wording and flow. Additionally, it is aimed to provide thorough training to enumerators prior to survey deployment, so they can conduct surveys in a conversational style. The investigators will employ measures such as data checks for straight lining and other signs of poor respondent engagement. The investigators will onboard reputable data collection staff who have a strong similar work experience.

6.4.1 High-frequency data quality checks and questionnaire review After data collection teams have submitted their data to the server, all questionnaires will be subject to high-frequency data checks. These data checks will identify any inconsistencies in the questionnaires. Field team members and/or interviews with a high rate of inconsistencies will be subject to closer review. The Data Analyst will run the high-frequency data checks using a defined set of criteria that may be updated during the data collection. Flagged questionnaires may be reviewed in full by study team members in order to identify issues. When these quality checks identify issues requiring field team re-training or disciplinary action, the analyst will notify the supervisor (field team lead). The supervisor will communicate the issue to the enumerator. If additional data collection is required (for example, a phone call to correct a small number of questions, or a re-interview in more extensive cases), the data collection firm will record this and share with the Surgo team.

6.4.2 Field supervision and spot checks Field teams will consist of supervisors and data collectors. Supervisors will be primarily responsible for conducting direct oversight of interviews, and offering support and correction as needed. Spot checks are unannounced visits by team leadership to observe an enumerator administer a questionnaire. Spot checks will be conducted by the team leaders and the study supervisors. Personel from the sponsor agency or someone whom the sponsor agency mandated for the quality assurance can also do the spot checks with the prior approval from household heads. During spot checks, the Survey Management team will look for enumerator competence and a thorough understanding of the questionnaire and how to administer it.

6.4.3 Back checks A subset of quantitative interviews - approximately 10% - will be subject to independent verification of key variables by an in-person or phone-based interview. These back-check interviews will allow for additional quality control and verification of responses.

8 ETHICAL CONSIDERATIONS 8.1 INFORMED CONSENT PROCESS This study will be submitted to the Ethics Review Committee (ERC) at AKU and National Bioethics Committee (NBC) Pakistan for ethical approval. All data collection will take place under informed consent. Respondents will be asked to provide verbal explanation of consent in the language they are most comfortable in and will be offered a summary sheet of the study and the details of the consent statement.

Quality control measures will be put into place to ensure that field team members follow the consent process according to these methods; these measures may include combined visit by the team leader or study supervisor for quality assurance. This can include voice recording with prior verbal approval from the study participant.

8.2 RISKS AND DISCOMFORTS Participants in this study may not experience potential physical risks and discomforts. However, discussing topics related to family planning and contraception can be sensitive and may evoke emotional distress or discomfort. There may also be concerns about privacy and confidentiality. To mitigate these risks, all participants will be oriented through the consent process about the importance of their participation and responses.

Inclusion Criteria:

• Married women aged 15-49
• Married women providing written informed consent to participate in the study
• One married woman from each selected household

Exclusion Criteria:

• Women below 15 years and above 49 years are not eligible to participate
• Women aged 15-18 will be excluded unless a parent, guardian, or spouse consents to her participation in the study,
• Participants who do not speak the languages in which the surveys and interviews are conducted may be excluded to ensure accurate communication and understanding of the study materials.
• Individuals with severe cognitive impairments or intellectual disabilities that could affect their ability to comprehend the study objectives or provide meaningful responses may be excluded.
• Participants who are unable or unwilling to provide informed consent for participation in the study will be excluded to uphold ethical standards and protect participant rights.