Predictors of Psychological Stress Among Street-Connected Children: A Multinomial Analysis of Risk and Protective Factors

Background

Street-connected children who live and work on the streets have been found to experience elevated levels of psychological stress (Maepa, 2021). Despite increasing recognition of their vulnerability, few empirical studies have explored how psychosocial risks, and protective factors relate to psychological stress within this population, particularly in West Africa. This study addresses this gap by identifying key risk and protective factors of psychological stress among street-connected children in Ghana.

Objective

The study aimed to examine associations between psychological stress and multiple risk and protective factors among street-connected children and to evaluate whether NGO service use, independently or in interaction with gender, is associated with stress level. The primary hypothesis tested was that there is no significant association between stress levels and NGO service usage among the study population.

Methods

Data were collected using a cross-sectional survey design with 273 street-connected children across three urban cities in Ghana. Of the children sampled, 1 in 2 were below age 14 years. Stress was measured using the Stress Numerical Rating Scale (SNRS-11) and categorized into four levels. Multinomial logistic regression was used to model predictors of stress level. Independent variables included gender, age, abuse on the street, hours of sleep, contact with family, and NGO service use etc. An interaction term between gender and NGO service use was also tested. We used Wald tests to assess the joint significance of NGO-related variables.

Results

The overall model was statistically significant (χ²(48) = 455.66, p < .001), with a Pseudo R² of 0.606. Hours of sleep was the most robust and consistent protective factor, with longer sleep duration significantly associated with decreased odds of being in the low (RRR = 0.19), moderate (RRR = 0.01), and severe (RRR = 0.003) stress categories compared to no stress (all p < .001). Street abuse significantly increased the odds of severe stress (RRR = 8.73, p = .049). Unexpectedly, very frequent family contact was associated with greater odds of both moderate (RRR = 4.89, p = .038) and severe stress (RRR = 6.75, p = .009. Female participants had significantly lower odds of reporting moderate (RRR = 0.003, p = .004) and severe stress (RRR = 0.009, p = .021). While the interaction between female gender and NGO service use was significantly associated with severe stress (RRR = 8.01, p = .043), Wald tests showed no joint significance of NGO service use (χ²(3) = 3.07, p = .381), thus supporting the null hypothesis.

Conclusions and Implications

The findings underscore the importance of meeting basic physiological needs, particularly adequate sleep, in reducing psychological stress of street children. Trauma exposure, particularly street abuse, significantly increases the likelihood of severe stress, and frequent family contact may reflect complex or distressing family dynamics. While overall NGO service use did not significantly predict stress levels, the gendered interaction points to the importance of evaluating how service delivery is experienced differently by boys and girls. Future studies are needed to explore the nature and quality of family relationships among street-connected children.