Diabetes self-management education interventions and self-management in low-resource settings; a mixed methods study.

INTRODUCTION: Diabetes is largely a self-managed disease; thus, care outcomes are closely linked to self-management behaviours. Structured self-management education (DSME) interventions are, however, largely unavailable in Africa. AIM: We sought to characterise DSME interventions in two urban low-resource primary settings; and to explore diabetes self-management knowledge and behaviours, of persons living with diabetes (PLD). RESEARCH DESIGN AND METHODS: A convergent parallel mixed-methods study was conducted between January and February 2021 in Accra, Ghana. The sampling methods used for selecting participants were total enumeration, consecutive sampling, purposive and judgemental sampling. Multivariable regression models were used to study the association between diabetes self-management knowledge and behaviours. We employed inductive content analysis of informants' experiences and context, to complement the quantitative findings. RESULTS: In total, 425 PLD (70.1% (n = 298) females, mean age 58 years (SD 12), with a mean blood glucose of 9.4 mmol/l (SD 6.4)) participated in the quantitative study. Two managers, five professionals, two diabetes experts and 16 PLD participated in in-depth interviews. Finally, 24 PLD were involved in four focus group discussions. The median diabetes self-management knowledge score was 40% ((IQR 20-60). For every one unit increase in diabetes self-management knowledge, there were corresponding increases in the diet (5%;[95% CI: 2%-9%, p<0.05]), exercise (5%; [95% CI:2%-8%, p<0.05]) and glucose monitoring (4%;[95% CI:2%-5%, p<0.05]) domains of the diabetes self-care activities scale respectively. The DSME interventions studied, were unstructured and limited by resources. Financial constraints, conflicting messages, beliefs, and stigma were the themes underpinning self-management behaviour. CONCLUSIONS: The DSME interventions studied were under-resourced, and unstructured. Diabetes self-management knowledge though limited, was associated with self-management behaviour. DSME interventions in low resource settings should be culturally tailored and should incorporate sessions on mitigating financial constraints. Future studies should focus on creating structured DSME interventions suited to resource-constrained settings.

Evaluation of a device to detect neonatal hypothermia in a clinical setting in Ghana.

Neonatal hypothermia poses an increased risk of infection, hypoglycemia, metabolic dysfunction, and mortality, particularly in preterm or low birthweight (LBW) infants. However, early detection of hypothermia and prompt thermoregulation can mitigate these effects thus, the need for continuous neonatal temperature monitoring. The BEMPU TempWatch is a small bracelet designed for continuous temperature monitoring for neonates. When the body temperature falls below 36.5˚C, the bracelet generates an alarm sound and flashes an orange light, indicating hypothermia. This study aimed to assess the validity of the BEMPU TempWatch in detecting hypothermia in a clinical setting in Ghana using sensitivity and specificity. Additionally, the study sought to identify factors associated with misclassification using logistic regression analysis. A standardized questionnaire collected information about the mother, pregnancy, delivery, and neonate. The BEMPU TempWatch was placed on the wrist of the neonate, and over a 24-hour follow-up period, a nurse took 4-hourly axillary temperature readings using a digital thermometer. Whenever the device's alarm sounded, a nurse immediately checked and recorded the axillary temperature, undertook necessary clinical actions, and rechecked after 30 minutes. Among the 249 neonates included in the study, 57.0% were female, 12.5% were extremely LBW, and 13.7% were extremely preterm. Based on 1,973 temperature readings, the sensitivity of the BEMPU TempWatch in detecting hypothermia was 67.8%, and the specificity was 95.9%. The sensitivity was lower among neonates being treated in incubators (58.4%) compared to those not (82.7%). Sensitivity was higher among neonates with LBW (1,500-2,500g) (73.5%) than very or extremely LBW neonates (<1,500g) (62.8%). The results showed that the BEMPU TempWatch had significantly fewer misclassifications among neonates who were not treated in an incubator, received only breastmilk, and were not born extremely preterm. Further studies are warranted to evaluate the effectiveness of the BEMPU TempWatch on neonatal health outcomes.