The neonatal mortality rate in Papua New Guinea is high, estimated at 24 per 1000 live births. The neonatal case fatality rate in newborns in provincial and referral hospitals was 5.9% in 2021. Deteriorating newborns can be difficult to identify. This observational study investigated the feasibility of using a neonatal colour coded observation and response chart to identify neonates at risk of deterioration and to promptly escalate care in the Special Care Nursery of Port Moresby General Hospital. The chart was adapted from the Plymouth Hospital Neonatal Early Warning System chart, and was used to collect data over 8 weeks between 1st May and 30th June 2022. One hundred and fifty seven (157) neonates were observed over the 72 hours following admission to the Special Care Nursery. Neonates were grouped into those that had triggers (vital signs that fell in the red zone) and a stable group (who had no observations in the red zone). Of the 157 patients recruited 72 (45.9%) were stable, and 85 (54.1%) had triggers that should prompt a response. Forty seven (55.3%) of the neonates in the trigger group had appropriate interventions. Neonates with observations in the red zone (triggers) were more likely to die in the first 72 hours compared with those with no triggers. Most of the nurse responded positively to the introduction of the chart. Whilst the Neonatal Early warning System is a tool that can be used to identify neonates at risk of clinical deterioration, proper training of its use and knowledge of and appropriate escalation of care are necessary to ensure its benefit

The lifespan of medical equipment depends on preventative maintenance. Properly functioning oxygen concentrators are the only practical sources of oxygen in many Low & Middle Income Countries and their use reduces mortality in hospitalised children. We provided 82 concentrators with pulse oximeters, split flow meters, oxygen tubing, and an oxygen analyser to 38 health facilities. Training and instructions on how to perform preventative maintenance were provided. The concentrators were monitored for three years after they were installed, by assessing the proportion of concentrators still producing optimal oxygen at greater than 85% purity, the proportion that underwent weekly maintenance checks, and the proportion that were faulty and repaired. A logbook for weekly documentation of performance, maintenance, faults and repairs, was employed. Faults were additionally identified by a biomedical engineer during the visits. Twenty nine oxygen concentrators underwent regular maintenance checks, 25 (86.2%) of which had a median of 30 (IQR: 9 - 65) checks. Twenty-four were functioning well throughout the three years. One concentrator was used for 23,807 hours before requiring repair. Fourteen (24%) of the 58 concentrators used at the start of the programme had problems, two were repaired, and 12 were replaced. Concentrator failure was mostly caused by excessive movement, dust, and leaking in the internal tubing. Routine preventative maintenance, thorough documentation of performance and reporting of problems, and having access to clinicians and a knowledgeable biomedical engineer are essential for oxygen concentrator longevity in health care facilities in low-resource settings.

Thiamine deficiency may be associated with severe acute malnutrition (SAM) and contribute to its clinical signs and complications. This hospital-based prospective study aimed to determine if thiamine deficiency is present in children with SAM in Port Moresby General Hospital (PMGH). Convenient sampling was used to select 208 children aged between 2 months and 13 years (median age 15 and IQR 11-32 months) presenting to the Children’s Emergency Department requiring venipuncture for routine testing who had additional blood taken for assay of thiamine levels. Nutritional status was determined using the WHO weight for height (WFH) z-scores. Details of socioeconomic background and diet were recorded. Thiamine levels were determined using enzyme-linked immunosorbent assay (ELISA) test kits. A total of 158 (76%) of the 208 samples were satisfactory for analysis; 86 (54%) from children with normal WFH, 26 (17%) from those with moderate malnutrition and 46 (29%) from those with severe malnutrition. Thiamine levels were normally distributed overall. The mean was 34.18  5.8 ng/ml. This was within the reference level of 16-48ng/ml. There was no statistically significant difference in levels between the normal, moderately and severely malnourished children (33.6  5.6, 35.3  5.7 and 34.4  5.7 ng/ml). The assay of thiamine levels in serum using ELISA is not the best method for determining thiamine deficiency and further studies using whole blood and high-performance liquid chromatography are needed

Problem: Emerging bacterial antimicrobial (antibiotic) resistance (AMR) is a global threat to human health. However, a majority of lower income countries do not have microbiological diagnostic testing for prompt, reliable confirmation of bloodstream infection and identification of AMR. Context: Clinicians in Pacific island nations are increasingly challenged by patients who have infection due to antimicrobialresistant bacteria. Treatment of infection remains empirical because of a lack of diagnostic testing capacity and may follow guidelines that were formulated without reference to local measures of AMR prevalence. There is limited understanding among clinicians of microbiology testing and test interpretation. Action: Examine the lessons learnt from pilot laboratory development programmes in two Pacific island nations that focused on establishing standard procedures for micrological diagnostics and antimicrobial susceptibility testing (AST) and on improving the training of clinicians to increase their use of laboratory services. Outcome: The pilot programmes addressed a range of logistical difficulties and evaluated two blood culture systems. They also examined and improved internal QC implementation and evaluated the prevalence of AMR. Discussion Continued development of microbiological diagnostic capability in the Pacific region is paramount. Pacific Island nations need to develop the capability of at least one central laboratory to culture AMR pathogens and subject them to quality-controlled AST or arrange for suitable referral to a nearby country.

Background: According to the Papua New Guinea (PNG) 2015 Annual Report on Child Morbidity and Mortality, neonatal conditions remained the second leading cause of hospital admission. We aimed to identify the common causes of neonatal admissions and deaths in the Eastern Highlands Provincial Hospital (EHPH) Special Care Nursery, and to identify interventions to improve neonatal outcomes in EHPH. Method: This was a retrospective descriptive study of neonatal admissions over a 5-year period: 2011 to 2015. Data collected from the ward admission register included demographic information, admission diagnoses, outcomes and causes of death where applicable. Results: From 2011 to 2015 there were 5177 neonatal admissions and 504 deaths with an overall case fatality rate of 9.7%. More than one admission diagnosis was recorded for some babies. The leading diagnoses were prolonged rupture of membranes (PROM) (25%), low birthweight (LBW) (20%), neonatal sepsis (NNS) (18%), meconium aspiration syndrome (MAS) (16%) and birth asphyxia (BA) (14%). The leading causes of death by admission diagnosis were LBW (259/1013, 26%), BA (141/746, 19%), MAS (80/852, 9%), NNS (79/915, 9%) and babies born before arrival (BBA) (61/306, 20%). 76% of the admissions were hospital-born babies, but 46% of deaths were in the 24% of babies born outside the hospital. The independent predictors of neonatal death were LBW (odds ratio [OR] 7.31), BA (OR 5.94), health centre birth (OR 3.07) and village birth (OR 2.18). Conclusion: LBW, BA and being born outside of a hospital were the strongest risk factors for neonatal death in the Eastern Highlands during this 5-year period. For PNG to achieve the Sustainable Development Goal of a neonatal mortality rate of 12 per 1000 live births, these 3 areas need to be addressed by multifaceted approaches within and outside the health sector.

Summary AIMS: This study aimed to evaluate a surveillance system for pigbel (enteritis necroticans), an enteric disease caused by Clostridium perfringens type C beta-toxin. Pigbel was formerly endemic in Papua New Guinea. Surveillance was based on a standardized clinical case definition, plus laboratory confirmation using a new assay for Clostridium enterotoxin that may be identified in intestinal secretions or stool. METHODS: Patients admitted to the Nazarene Hospital in the Jiwaka Province were eligible for inclusion if they had severe abdominal pain of less than 2 weeks’ duration. A standardized case definition of pigbel was used to distinguish cases of acute abdominal pain very likely to be due to pigbel from cases likely to be accounted for by other diagnoses. Children enrolled in the study were tested with a new ELISA (enzyme-linked immunosorbent assay) for Clostridium perfringens beta-toxin. The number and types of samples tested depended on the clinical features and illness severity, and included gastric aspirate, stool and intestinal fluid if surgery was required. RESULTS: Between October 2012 and January 2015 105 patients with acute abdominal pain were admitted. The median age (interquartile range) was 5 (2- 6) years. 48 children met the standardized criteria for the clinical diagnosis of pigbel. 34 of 67 children tested with ELISA fulfilled the criteria for clinical pigbel, of whom 12 tested positive and 22 negative. The sensitivity of the ELISA test against the clinical diagnosis was therefore 35%. The test was positive in 4 children with abdominal pain who did not fulfil all the clinical criteria for pigbel, suggesting that they had a mild form of the disease. There were 4 surgically confirmed patients who were ELISA negative. In total, laboratory or surgically confirmed pigbel was found in 20 patients, but it is likely that there were others with pigbel among the 105 patients with acute abdomen. 5 of the study children died. CONCLUSIONS: The ELISA test for Cl. perfringens beta-toxin can be a useful confirmatory test and can help to validate the clinical impression of pigbel. However, the sensitivity is low, even in children who are likely to have pigbel, and therefore a negative test cannot rule out the diagnosis. Because the differential diagnosis of the clinical features of pigbel is broad, having the combination of clinical and laboratory-based surveillance for pigbel would help monitor the true burden of disease, and could be used to guide public health interventions.

Child adoption is common in Papua New Guinea (PNG), but little is known about current practices. This study describes the adoption and feeding practices among children up to 3 years of age, identified by both community-based and health facility recruitment in the Southern Highlands of PNG over a 6-month period in 2017. 85 adopted children were identified: 61 (72%) had been customarily or informally adopted and 24 babies (28%) had been sold for cash. Most (93%) were adopted at less than 1 year of age, and twothirds in the neonatal period. 73 mothers (86%) had poor knowledge of the adoption laws and processes, and 56 (66%) had poor knowledge of infant feeding practices. 8% had moderate and 9% had severe malnutrition. The common reasons for adoption were: family breakdown because of divorce, separation or the practice of polygamy; mothers who had too many children or wanted to adopt children; teenage pregnancies; and de facto relationships. 56 of the adopted children were weaned before 6 months of age, and 23 before 4 months of age, most onto a diet consisting of a limited range of fruits and vegetables. Ensuring the rights of adopted children to legal protection, optimal nutrition, care and education is essential. Knowledge of and adherence to customary adoption laws, involvement of social welfare, and infant and young child feeding practices all need to be strengthened in communities, especially in the rural areas.

After 20 years of in-service training in child health, involving many different WHO and UNICEF courses: the Integrated Management of Childhood Illness (IMCI), Infant and Young Child Feeding (IYCF), Severe acute malnutrition (SAM), Early Essential Newborn Care (EENC) and many others, it is time to consider what model of training is efficient and sustainable. Such a training model will need to address the complexities of child health in Papua New Guinea (PNG) in the era of the Sustainable Development Goals (SDG) and provide health workers with a program of continuing professional development (CPD). Efficient models of training will have a common framework and integrate the best of these individual courses, and involve modern information technology to improve efficiency, be easily updatable and enlarge access. Training should be flexible and modular so that it can be delivered as an in-service course, a program of CPD within a hospital, or as self-learning. It should involve the principles of adult learning, enabling health workers to build on their existing knowledge and skills, to learn how to use standard up-to-date texts and technical resources in everyday clinical practice, and to understand where and how to access quality, credible health information. In-service training and CPD should involve and support PNG’s schools and colleges of nursing, HEOs and community health worker training, with the Paediatric Society, the School of Medicine and Health Sciences (SMHS) University of PNG, and the National Department of Health (NDoH) taking the lead. A course that integrates the best of the existing WHO and UNICEF courses could form the basis of a post-graduate child health nursing diploma, which could be established in more provinces to address the shortage of paediatric nurses

Background: Identifying the causes of childhood meningitis is difficult. Conventional diagnostic tests (culture, bacterial latex and Gram staining) have limitations, especially in settings where many children receive antibiotics prior to presentation. A point-of-care test called Binax NOW detects meningitis due to 'Streptococcus pneumoniae' in 15 minutes and is not affected by pre-test antibiotic use.

Methods: A prospective study was conducted among children with suspected bacterial meningitis at Angau Memorial General Hospital to evaluate the usefulness of the Binax NOW 'S. pneumoniae' test in comparison with conventional tests: cerebrospinal fluid (CSF) bacterial culture, Gram stain and latex agglutination. Latex antigen testing was done for 'S. pneumoniae, Haemophilus influenzae and Neisseria meningitidis'. We analysed the CSF of all children who had a lumbar puncture done for clinical suspicion of meningitis. FINDINGS: 132 children were enrolled in the study, of which 3 were excluded because of insufficient CSF sample to do the Binax NOW test. 5 CSFs were culture positive, all for 'S. pneumoniae'. 13 (10%) of 129 CSF specimens had organisms seen with Gram staining; 7 had Gram-positive cocci and 6 showed Gram-negative bacteria. Latex antigens were positive in 20 cases: for S. pneumoniae (11), 'H. influenzae' (8), 'N. meningitidis' (1). Using the 3 conventional tests combined (culture, Gram stain and antigens) 14 cases of 'S. pneumoniae' meningitis were detected. Binax NOW was positive for 'S. pneumoniae' in 19 cases (15% of meningitis cases): the 14 samples positive by conventional methods and a further 5 cases that were not detected by conventional methods.

Conclusion: The Binax NOW test increases the diagnostic yield for pneumococcal meningitis on CSF. This may be important in surveillance for the effectiveness of pneumococcal conjugate vaccine introduced in Papua New Guinea (PNG) in 2014, and in clinical diagnosis. 'H. influenzae' is rarely identified on culture in PNG provincial hospital laboratories, so latex antigen testing is still needed for the accurate diagnosis of 'Haemophilus' meningitis and monitoring of the effectiveness of 'Haemophilus influenzae' type b vaccine.
Streptococcus pneumoniae