7-Day vs 1-Day Packaging for Small Quantity Lipid-Based Nutrient Supplements in Ghana (SQLNS:7D-1D Comparison)

Background and Rationale: Small-quantity lipid-based nutrient supplements (SQ-LNS) have demonstrated effectiveness in reducing child undernutrition, including stunting, wasting, and anemia, particularly among children 6-24 months of age in low- and middle-income countries. Despite strong evidence, uptake and adherence in real-world settings remain suboptimal. Packaging format may influence how caregivers understand, use, and integrate SQ-LNS into daily routines, and it also determines packaging cost and distribution burden.

Study Objectives: Primary Objective: To determine whether adherence to SQ-LNS delivered in 7-day resealable pouches is non-inferior to adherence to SQ-LNS delivered in 1-day sachets. Secondary Objectives: To assess caregiver acceptability, preferences, ease of use, and perceptions of both packaging formats; to assess observed usability at the point of dispensing; and to explore implementation facilitators and barriers to effective delivery and use of SQ-LNS within growth monitoring and promotion programs.

Study Design: This is a cluster-randomized, two-period crossover trial embedded within routine child health services at up to 16 public health facilities in the Kumbungu and Tolon districts of Northern Ghana. The trial follows a convergent mixed-methods implementation research design. Clusters are health facilities, each randomized to one of two packaging sequences. Each facility delivers one packaging format for one month, then switches to the alternate format for a second month. Participants are caregivers of children aged 6-24 months attending growth monitoring and promotion (GMP) services during the intervention periods.

Randomization: Facilities are matched by facility type (hospital, health centre, or CHPS compound), location, and average monthly GMP attendance, then randomized to one of two sequences: Sequence A, 1-day sachet followed by 7-day pouch; Sequence B, 7-day pouch followed by 1-day sachet. Randomization is at facility level and is unblinded.

Sample Size: The primary comparison is a non-inferiority test with a pre-specified margin of 5.0 percentage points of adherence. Assuming a within-subject adherence standard deviation of 20%, a within-subject correlation of 0.60, a one-sided significance level of 0.025, and 80% power, the primary comparison is powered at over 99% at the planned sample of 505 caregiver-child dyads. The enrollment target of 505 is set by the precision required for the caregiver-level acceptability and preference outcomes, estimated to a 95% confidence interval half-width of 5 percentage points, which requires 385 participants with complete endline data and approximately 480 enrolled after allowing for 20% loss to follow-up. Dyads are allocated across facilities in proportion to average monthly GMP attendance rather than equally, and are enrolled consecutively during routine GMP sessions until each facility's target is reached.

Outcomes: The primary outcome is weight-based adherence, calculated as 100 x (grams dispensed - grams returned unconsumed) / grams dispensed, net of packaging tare, collected for every enrolled dyad in each period. A weight-based definition is used because the 7-day pouch has no countable dose units, making weight the only metric measurable identically in both arms. Count-based consumption in the sachet arm and self-reported adherence are retained as supportive measures. Secondary outcomes include caregiver acceptability and preference, dispensing accuracy within 2 g of the 20 g target, package opening without a tool or assistance, and qualitative themes relating to packaging, feeding practices, and caregiver experience.

Data Collection: Quantitative data are collected through caregiver surveys at baseline, midline, and endline, and through weighing of returned product. Returned product is weighed at the first contact point that reaches the caregiver, following a fixed sequence: the next scheduled GMP clinic visit; if not captured there, a community outreach growth-monitoring event; and if still not captured, a home visit. Usability is assessed in two components: a brief facility-based observation of package opening and dose dispensing for all dyads at midline and endline, and a home observation of feeding, storage, and household environment for a randomly selected sub-sample of approximately 35 dyads stratified by facility type. Routine growth-monitoring measurements are abstracted from the child's Maternal and Child Health Record Book and from clinic registers; no anthropometric measurements are taken by study staff. Qualitative data are collected after the second crossover period through approximately 20-22 in-depth interviews with caregivers, frontline health workers, and program managers, and up to 8 focus group discussions with caregivers who completed both phases.

Analysis Plan: The primary outcome is analyzed as a continuous variable using a linear mixed model with fixed effects for packaging format, period, and sequence, and random intercepts for child and facility. Non-inferiority of the 7-day pouch to the 1-day sachet is concluded if the lower bound of the two-sided 95% confidence interval for the mean difference in adherence lies above -5.0 percentage points. The analysis is conducted in both per-protocol and intention-to-treat populations, with non-inferiority required in both, and period and carryover effects assessed. The intention-to-treat analysis uses multiple imputation as the primary approach, with a conservative sensitivity analysis and a tipping-point analysis. Secondary outcomes are analyzed using linear mixed models and standard two-sided comparisons; the non-inferiority framing applies only to the primary adherence outcome. Pre-specified subgroup analyses examine variation by facility type, socioeconomic status, and caregiver education. Qualitative data are coded and analyzed using thematic content analysis in NVivo, with codes developed both deductively from study objectives and inductively from emerging data. Quantitative and qualitative findings are triangulated to inform implementation recommendations.

Ethical Considerations: The study has been approved by the Boston University Charles River Campus IRB (protocol 8287E) and by the Ghana Health Service Ethics Review Committee (GHS ERC 006/12/25). Written informed consent is obtained from all participating caregivers before any study activity. The study poses minimal risk and is embedded in routine health services.

Dissemination: Findings will be shared with the Ghana Health Service, implementing partners, and international stakeholders, and will be submitted for publication in peer-reviewed journals.