The Institute for Implementation Science Scholars (IS-2) is a dissemination and implementation (D&I) science training and mentoring program. A key component of IS-2 is collaborating and networking. To build knowledge on effective networking and mentoring, this study sought to 1) conduct a social network analysis to determine whether underrepresented scholars have equivalent levels of connection and 2) gain insights into the differences in networking among racial/ethnic subgroups of scholars.

Social network survey data were used to select participants based on number of collaborative connections (highest, lowest) and racial/ ethnic category (underrepresented, not underrepresented). Interviews were recorded, transcribed, and coded using an iterative process.

The sample consisted of eight highly networked scholars, eight less networked scholars, seven from underrepresented racial and ethnic groups, and nine from not underrepresented groups. Qualitative data showed a lack of connection, reluctance to network, and systematic issues including institutional biases as possible drivers of group differences. In addition, scholars provided suggestions on how to overcome barriers to networking and provided insights into how IS-2 has impacted their D&I research and knowledge.

Underrepresented scholars have fewer network contacts than not underrepresented scholars in the IS-2 training program. It is imperative for leadership to be intentional with mentorship pairing, especially for underrepresented scholars. Future research might include interviews with program leaders to understand how network pairings are built to improve the mentorship experience.

We aimed to assess the maternal and family determinants of four anthropometric typologies at the household level in Colombia for the years 2000, 2005 and 2010.

We classified children <5 years old according to height-for-age Z-score (<−2) and BMI-for-age Z-score (>2) to assess stunting and overweight/obesity, respectively; mothers were categorized according to BMI to assess underweight (<18·5 kg/m2) and overweight/obesity (≥25·0 kg/m2). At the household level, we established four final anthropometric typologies: normal, underweight, overweight and dual-burden households. Separate polytomous logistic regression models for each of the surveyed years were developed to examine several maternal and familial determinants of the different anthropometric typologies.

National and sub-regional (urban and rural) representative samples from Colombia, South America.

Drawing on data from three waves of Colombia’s Demographic and Health Survey/Encuesta Nacional de Salud (DHS/ENDS), we examined individual and household information from mothers (18–49 years) and their children (birth–5 years).

Higher parity was associated with an increased likelihood of overweight and dual burden. Higher levels of maternal education were correlated with lower prevalence of overweight, underweight and dual burden of malnutrition in all data collection waves. In 2010, participation in nutrition programmes for children <5 years, being an indigenous household, food purchase decisions by the mother and food security classification were also associated with the four anthropometric typologies.

Results suggest that maternal and family correlates of certain anthropometric typologies at the household level may be used to better frame policies aimed at improving social conditions and nutrition outcomes.