STRIVE PNG: using a partnership-based approach in implementation research to strengthen surveillance and health systems in Papua New Guinea.

Successful implementation research requires effective and equitable relationships between policy-makers, researchers and implementers to effect evidence-based systems change. However, mainstream research grant models between Global North and Global South institutions often (unintentionally) reinforce power imbalances between partners, which result in missed opportunities for knowledge and learning exchange between policy-makers, researchers and implementers.This case study, centred on the STRIVE PNG project, describes how a partnership-based approach has been used to establish, maintain and review effective and equitable relationships between 13 partner organizations (independent research institutes, government health agencies and public health laboratories) to strengthen surveillance and health systems in Papua New Guinea (PNG). We provide an overview of key terms (with supporting conceptual frameworks), describe selected partnership processes and tools used within the project, and share observations regarding early outcomes achieved through this approach.

Surveillance of molecular markers of Plasmodium falciparum artemisinin resistance (kelch13 mutations) in Papua New Guinea between 2016 and 2018.

Plasmodium falciparum resistance to artemisinin-based combination therapy (ACT) is a global threat to malaria control and elimination efforts. Mutations in the P. falciparum kelch13 gene (Pfk13) that are associated with delayed parasite clearance have emerged on the Thai-Cambodian border since 2008. There is growing evidence of widespread Pfk13 mutations throughout South-East Asia and they have independently emerged in other endemic regions. In Papua New Guinea (PNG), Pfk13 "C580Y" mutant parasites with reduced in vitro sensitivity to artemisinin have been isolated in Wewak, a port town in East Sepik Province. However, the extent of any local spread of these mutant parasites in other parts of PNG is unknown. We investigated the prevalence of Pfk13 mutations in multiple malaria-endemic regions of PNG. P. falciparum isolates (n = 1152) collected between 2016 and 2018 and assessed for Pfk13 variation by sequencing. Of 663 high quality Pfk13 sequences a total of five variants were identified. They included C580Y, a mutation at a previously documented polymorphic locus: N499K, and three previously undescribed mutations: R471C, K586E and Y635C. All variants were found in single isolates, indicating that these Pfk13 mutations were rare in the areas surveyed. Notably, C580Y was absent from Maprik district, which neighbours Wewak where C580Y mutant parasites were previously identified. The single C580Y isolate was found in the port town of Lae, Morobe Province, a potential entry site for the importation of drug resistant parasites into PNG. Although sample size in this location was small (n = 5), our identification of a C580Y mutant in this second location is concerning, highlighting the urgent need for further surveillance in Lae. Other Pfk13 mutants were rare in PNG between 2016 and 2018. Continued surveillance for molecular markers of drug resistance is critically important to inform malaria control in PNG.

Medical Students at the School of Medicine and Health Sciences, University of Papua New Guinea: Predictors of Performance and Student Backgrounds.

BACKGROUND: Papua New Guinea, a lower middle income country with a population of around 8.5 million, the majority of whom live in rural areas, produces far fewer than the number of medical graduates required to meet the WHO-recommended doctor/population ratio. The School of Medicine and Health Sciences is under pressure to increase its output and ensure the graduates are able to function in rural settings. Through two studies, we aimed to determine the predictors of student performance and their socioeconomic and educational background to assist in determining admission policies and improve completion rates. METHODS: A retrospective study analysed data relating to student performance from six annual cohorts. A cross-sectional study among currently enrolled students sought information about their socioeconomic and educational background. RESULTS: Of the 300 students enrolled in the six cohorts, 176 (59%) completed the programme in the scheduled 4 years. There were no differences in completion rates by gender or route of entry to the programme. Grade point average at medical school entry predicted academic performance. Sixty-four per cent of the students who failed to complete in four years attributed their poor academic performance to social issues. Overall attrition was only 8%. Seventy-six per cent (162/214) of the enrolled students completed the cross-sectional survey. Most (79%) of students' fathers and 58% of mothers had postsecondary education. Seventy-three per cent of respondents indicated that they had been to preschool or elementary school. Thirty-six per cent had attended primary school in a village or government/mission station. Just over half (53%) of the students indicated that English had been the language most used in primary school. Males were more likely to have made a specific career choice than females. The majority (141/162, 88%) of the students indicated that they had experienced some academic difficulty during the years. CONCLUSION: Prior academic performance predicted timely completion of the MBBS programme. Just over a third of students had attended rural village primary schools. Social and domestic issues were common and adversely affected academic performance.

Validation of a Dried Blood Spot Ceftriaxone Assay in Papua New Guinean Children with Severe Bacterial Infections.

Dried blood spot (DBS) antibiotic assays can facilitate pharmacokinetic (PK) studies in situations where venous blood sampling is logistically and/or ethically challenging. In this study, we aimed to demonstrate the validity of a DBS ceftriaxone assay in a PK study of children with severe illness from Papua New Guinea (PNG), a setting in which health care resources are limited and anemia is common. Using a previously validated liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay, serial plasma and DBS ceftriaxone concentrations were measured in PNG children aged 5 to 10 years with acute bacterial meningitis or severe pneumonia. The concentration-time data were incorporated into population PK models. Ten children were recruited with an admission hematocrit of 0.22 to 0.52. Raw data demonstrated good correlation between plasma and DBS concentrations (Spearman's rank correlation coefficient [rs] = 0.94 [95% confidence interval, 0.91 to 0.97], P < 0.0001). A marked systematic hematocrit bias was observed, with lower hematocrits resulting in underestimation of DBS-predicted plasma concentration. After adjustment for red cell partitioning and hematocrit bias, a population PK model comparing plasma and DBS-predicted plasma concentrations did not differ in terms of key PK parameters, including clearance, volume of distribution, and residual variability. The performance of the ceftriaxone DBS assay is robust and provides reassurance that this platform can be used as a surrogate for plasma concentrations to provide valid PK and PK/pharmacodynamic studies of severely unwell children hospitalized in a resource-limited setting. It highlights the importance of hematocrit bias in validation studies of DBS assays.

Evaluation of Xpert MTB/RIF assay in children with presumed pulmonary tuberculosis in Papua New Guinea.

BACKGROUND: The Gene Xpert MTB/ RIF assay (Xpert) is used for rapid, simultaneous detection of Mycobacterium tuberculosis (MTB) and rifampicin resistance. This study examined the accuracy of Xpert in children with suspected pulmonary tuberculosis (PTB). METHODS: Children admitted to Port Moresby General Hospital with suspected PTB were prospectively enrolled between September 2014 and March 2015. They were classified into probable, possible and TB-unlikely groups. Sputum or gastric aspirates were tested by Xpert and smear microscopy; mycobacterial culture was undertaken on a subset. Children were diagnosed with TB on the basis of standard criteria which were used as the primary reference standard. Xpert, smear for acid-fast bacilli (AFB) and the Edwards TB score were compared with the primary reference standard. RESULTS: A total of 93 children ≤14 years with suspected PTB were enrolled; 67 (72%) were classified as probable, 21 (22%) possible and 5 (5.4%) TB-unlikely. Eighty were treated for TB based on the primary reference standard. Xpert was positive in 26/93 (28%) MTB cases overall, including 22/67 (33%) with probable TB and 4/21 (19%) with possible TB. Three (13%) samples identified rifampicin resistance. Xpert confirmed more cases of TB than AFB smear (26 vs 13, p = 0.019). The sensitivity of Xpert, AFB smear and an Edwards TB score of ≥7 was 31% (25/80), 16% (13/80) and 90% (72/80), respectively, and the specificity was 92% (12/13), 100% (13/13) and 31% (4/13), respectively, when compared with the primary reference standard. CONCLUSION: Xpert sensitivity is sub-optimal and cannot be relied upon for diagnosing TB, although a positive result is confirmatory. A detailed history and examination, standardised clinical criteria, radiographs and available tests remain the most appropriate way of diagnosing TB in children in resource-limited countries. Xpert helps confirm PTB better than AFB smear, and identifies rifampicin resistance. Practical guidelines should be used to identify children who will benefit from an Xpert assay.

Severe malnutrition in children in Papua New Guinea: effect of a multi-faceted intervention to improve quality of care and nutritional outcomes.

BACKGROUND AND AIMS: Severe malnutrition remains a major problem in Papua New Guinea; it is associated with 11% of paediatric hospital admissions and 33% of all child deaths, with a case fatality rate around 20%. This article aims to evaluate the effectiveness of a multi-faceted intervention for improving care for children with severe malnutrition. METHODS: Severe malnutrition was defined as weight-for-age (WFA) <-3 Z-scores with severe wasting or mid upper arm circumference <115 mm or generalised oedema owing to malnutrition. The intervention included training for health-care workers on WHO guidelines for severe malnutrition, ward-round checklists, posters and support for nurses to provide better patient nutrition. Three point prevalence surveys were conducted; one before the intervention and two afterwards at 3-month intervals. The main outcomes were weight change since admission, energy intake and the proportion of the calculated required energy intake in the previous 24 hours. Each stage of the WHO guidelines for severe malnutrition management was assessed for adherence. RESULTS: There were significant improvements in the WHO steps for the management of severe malnutrition. At pre-intervention baseline, children received a median of 356 ml/day (IQR 178-450): 31% (95% CI 21-48) of their estimated daily energy requirements for weight. In the first follow-up survey, children received a median of 820 (IQR 600-1110) ml/day: 98% (95% CI 67-100) of daily energy requirements; and in the second follow-up survey they received 780 (IQR 480-900) ml/day: 86% (95% CI 46-100%) of daily requirement (P<0.001 both for volume received and percentage of energy requirements). Median weight gain prior to the intervention was 1.55 g/kg/day (IQR -4.3-6.0) which increased to 5.56 g/kg/day (IQR -3.7-12.0) and 10.19 g/kg/day (IQR 0-16.0) in the first and second follow-up surveys, respectively (P=0.013). CONCLUSION: Implementation of a multi-faceted intervention to improve the management of children with severe malnutrition was associated with improved quality of care and improved weight gain.

Risk factors for malnutrition in children at Port Moresby General Hospital, Papua New Guinea: a case-control study.

Fifty children admitted for malnutrition were age matched with 50 admitted for other reasons. These children were more likely to be female (p = 0.003), born low birth weight (p = 0.02), after a short birth interval (p = 0.014) and to be incompletely vaccinated (p < 0.001) than control children, and to be living in rural villages or settlement housing (p < 0.001) with inadequate water supply (p < 0.001) and sanitation (p = 0.037), with overcrowding (p = 0.016) and low household income (p = 0.04). Their parents were more likely to have had no or only rudimentary education than parents of control children [Odds ratio (OR) 3.58 for mothers, 4.12 for fathers]. Parental consumption of alcohol as well as smoking in the mother was more common in the malnourished children. Running water in the house was an independent protective factor (OR 0.23) and the fathers' poor employment status (OR 4.12) an independent risk factor. The solution to malnutrition involves improving community understanding of nutrition and in reducing social inequalities.

Evaluation of an interferon-gamma release assay in children with suspected tuberculosis in Papua New Guinea.

There are few data from tuberculosis (TB) endemic settings of the performance and outcome predictors of the QuantiFERON-TB Gold in Tube assay (QFT) in children with suspected TB. A prospective cross-sectional study was conducted in Papua New Guinea children with suspected TB evaluated at Port Moresby General Hospital (Port Moresby, Papua New Guinea). Two hundred sixteen children were enrolled including 106 probable TB, 87 possible TB and 23 without TB. Concordance between QFT and tuberculin skin test results was 86% (P < 0.001, κ = 0.70). QFT was significantly more likely to be positive than tuberculin skin test, overall and within the probable or possible TB categories, with no difference in prevalence of positivity between these 2 categories. The role of QFT in supporting the clinical diagnosis of TB in endemic settings, where resources are limited, remains uncertain especially as cost and technical requirements remain considerable.

Papua New Guinea: real progress towards MDG 4 and real challenges.

With a mortality rate in the under-5 s of 93 per 1000 live births reported in the 1996 Demographic and Health Survey (DHS), Papua New Guinea (PNG) was at the time one of only four countries with stalled progress in child survival, and seemed destined to fail its national Millennium Development Goal (MDG) 4 target. However, accurate estimates have shown reductions in under-5 and infant mortality rates of 19% and 17% respectively, over 10 years from 1996 to 2006. In that period PNG adopted an integrated and coordinated approach to child health that includes all the essential interventions outlined in the Lancet's child survival series, under a framework consistent with the Western Pacific Regional Child Survival Strategy, associated with significant improvements in leadership and coordination of child health services by paediatricians at the provincial and national level. The reduction in child mortality since the mid-1990s is strong encouragement that such an approach can translate to real improvements. This paper outlines the recent advances in child health in PNG, identifying successful areas, and the challenges that lie ahead. There has been increased immunization coverage, introduction of vitamin A supplementation, bed-nets to prevent malaria, interventions to reduce mortality from acute respiratory infection, and improvements in the education of girls. These and improved leadership and coordination help to explain the recent significant gains in child survival.

The management of paediatric patients in a general Emergency Department in Papua New Guinea.

Children less than 13 years of age account for 27% of the case mix at the Emergency Department (ED) of the Port Moresby General Hospital (PMGH). The ED is busy, usually overcrowded, understaffed and under-equipped, resulting in less than optimal patient management. Children are a highly vulnerable group of patients and have the potential to deteriorate rapidly. This prospective descriptive study aimed to assess the adequacy of management of children presenting to the ED between 1600 and 0800 hours. A standardized and individually administered questionnaire was used to assess the management of 107 children. The median age was 13 months, interquartile range 6-36 months, with a male to female ratio of 1.5:1. The most frequent diagnoses were pneumonia/bronchiolitis, diarrhoea, malaria, asthma and febrile convulsions. Three-quarters of the sample were classified as being triage 1 and 2, ie, requiring either immediate life-saving treatment or treatment within 30 minutes to an hour of presentation. Median and interquartile ranges for time from arrival to assessment were 60 (15-110) minutes for triage 1, and 60 (30-121) minutes for triage 2 patients. Time from assessment to management was 5 (5-45) minutes for triage 1 and 40 (30-63) minutes for triage 2 patients. Treatment instituted was appropriate in 93% of cases but the drug dosage was incorrect in 26%. 49 children (46%) were admitted to the wards either directly or following further observation in the ED or Children's Outpatient Department, the rest being treated and discharged, except for one child with probable septicaemia who died following a prolonged and unattended wait in the ED. Management was assessed as adequate in only 40% of cases. The major causes of inadequate management were delayed treatment, under- or over-dosing, under- or over-treatment, omission of appropriate investigations, misdiagnosis and failure of judicious consultation with the paediatric team. Many patients were nursed on the floor. Recommendations emanating from the study include ensuring adequate staffing levels and the training of all staff working in ED in the rapid identification of sick children to improve triage and subsequent management.

Immune response to measles vaccine in 6 month old infants in Papua New Guinea.

OBJECTIVE: To assess the efficacy of the current measles immunization schedule in Papua New Guinea, which is to give the first dose at 6 months of age and the second at 9 months. METHODS: Humoral immune response study of 140 Papua New Guinean infants at 6 months of age, measuring measles IgG antibodies by enzyme immunoassay before and 85 days after the 6-month dose of measles vaccine. RESULTS: After vaccination at 6 months, 35.7% of infants developed a level of measles antibodies consistent with protection (IgG >330 IU/ml); 17.7% had an antibody response (150-330 IU/ml) that is likely to afford some protection; 46.8% had no detectable antibody response (IgG <150 IU/ml). Among 53 infants with no antibody response, 37 (69.5%) developed an antibody response, while 42.4% (37/87) of those with maternal antibodies sero-converted (P = 0.002). CONCLUSIONS: Antibody response to measles vaccine was lower than expected at 6 months. While the presence of maternally derived antibodies accounted for some of the limited seroconversion in young infants, other factors are involved. Issues to be considered in determining the value of the first dose of measles vaccination in mid infancy in poor countries are complex and antibody responses are only one factor. Others, such as cell mediated immune responses, the non-specific protective effect of measles vaccine in preventing illness and death and the practicalities of uptake of vaccines at different ages, are also important.

Reference values for pulse oximetry in healthy children in coastal Papua New Guinea.

We expected oxygen saturation (SpO2) in children in coastal Papua New Guinea (PNG) to be higher than in PNG highlands children. Therefore, SpO2 was documented to determine the reference values of SpO2 in neonates and young children; 149 healthy neonates and 100 healthy infants and children < 5 years old were studied in Port Moresby. SpO2 ranged from 93% to 100% in both groups. The median SpO2 in neonates was 97% (CI 96.9-97.4) and in young children 98% (CI 97.5-98.0). We recommend 93% as the cutoff for administering oxygen to children under 5 years old in coastal PNG.

Can clinical signs predict hypoxaemia in Papua New Guinean children with moderate and severe pneumonia?

Pulse oximetry was performed on 77 children admitted with acute lower respiratory tract infections (ALRI) to the children's ward in Port Moresby General Hospital, Papua New Guinea over a 4-month period in 2002. Clinical findings were correlated with different levels of hypoxaemia, <93%, <90% and <85%. Cyanosis, head nodding and drowsiness were good predictors of hypoxia but lacked sensitivity. Decisions to use oxygen based on these signs would therefore result in a significant number of children with hypoxia not receiving oxygen. Pulse oximetry is the best indicator of hypoxaemia in children with ALRI and, although relatively expensive, its use might be cost-effective in controlling oxygen requirements.

The medium-to-long-term outcome of Papua New Guinean children after cardiac surgery.

This study reports the medium-to-long-term outcome in Papua New Guinean (PNG) children selected to undergo cardiac surgery at the Royal Alexandra Hospital for Children in Sydney, Australia between 1978 and 1994. Follow-up ranged from 4 to 20 (median 11) years. The cohort comprised 125 children who had surgery and 31 who were initially selected in PNG for surgery but who on further investigation were found to be unsuitable. Through strenuous attempts, local health workers, the media and village and church leaders traced 122 (98%) of the operated and 29 (94%) of the non-operated children. One of the operated children and six of the non-operated children had died, giving respective survival rates among those traced of 99% and 79%. Altogether, 106 (88%) of the 121 operated and 20 (87%) of the 23 non-operated survivors were reviewed. Ninety-nine (93%) of the surgical patients were asymptomatic and all fulfilled the New York Heart Association criteria (NYHAC) class I or II. Mild pulmonary hypertension or residual defects of no haemodynamic significance were present in 47 (44%). In contrast, all 11 survivors from the 18 children originally classified as having inoperable lesions were symptomatic, all in NYHAC classes III or IV, six were on cardiac medication and four had been admitted at least once in the previous year. Ninety-two of 96 (96%) of the surgical group had a normal exercise test and 75 of 96 (78%) had normal chest X-rays. Thirty-nine of 99 had a normal electrocardiogram whilst the remainder had changes related to the underlying lesion and the surgery performed. This study shows that the PNG children who had cardiac surgery at RAHC between 1978 and 1994 had good medium-to-long-term survival.