Endemic Burkitt lymphoma: a 28-day treatment schedule with cyclophosphamide and intrathecal methotrexate.

BACKGROUND: Endemic Burkitt lymphoma (eBL) is the most common childhood cancer in equatorial Africa and there is a need for affordable, effective treatment. AIM: To record the morbidity of treatment and event-free survival after 1 year using relatively high doses of cyclophosphamide at short intervals combined with intrathecal methotrexate. METHODS: Forty consecutive patients with a mean age of 6.9 (range 2-15) years were treated at Queen Elizabeth Central Hospital, Blantyre between 10th April and 17th November 2006. The initial diagnosis was made clinically and confirmed by fine-needle aspiration in 73%. Abdominal ultrasound, bone marrow aspirate and CSF analysis were undertaken routinely. Chemotherapy consisted of cyclophosphamide, 40 mg/kg on day 1 and 60 mg/kg on days 8, 18 and 28. Intrathecal methotrexate 12.5 mg and hydrocortisone 12.5 mg were administered on days 1, 8, 18 and 28. Allopurinol was commenced before chemotherapy, and a high urinary output was maintained to prevent tumour lysis. RESULTS: St Jude stage distribution was stage I, 1; II, 9; III, 24; and IV, 6. An equal number (70%) presented with abdominal and facial disease, and 15% with paraplegia. Twenty patients (50%) were below the 5th NCHS centile for weight-for-age. Two patients died during treatment, three had chemotherapy-resistant disease and 35 (88%) achieved complete clinical remission by day 28. Sixteen required antibiotic treatment for presumed infection and nine received a blood transfusion. Relapse occurred in 16 patients after 65-311 days (median 137). Nineteen patients (48%) have been in continued remission for 265-670 days (median 454). CONCLUSION: This short, inexpensive treatment schedule (<50 US$) cured almost 50% of eBL patients in a setting of very limited resources.

Rescue chemotherapy for patients with resistant or relapsed endemic Burkitt's lymphoma.

Patients with endemic Burkitt's lymphoma who failed primary treatment with the Malawi 2002 or 2003 Burkitt's lymphoma treatment protocols, consisting of high frequency cyclophosphamide 40 mg/kg and intrathecal methotrexate, were offered rescue chemotherapy. Twenty-eight patients (14 boys and 14 girls; age range 3-13 years) with resistant disease (n=8) or relapse (n=20) presented to the Queen Elizabeth Central Hospital, Blantyre, Malawi. Treatment consisted of cyclophosphamide 60 mg/kg and vincristine 1.5 mg/m(2) i.v. on Days 1, 8 and 15, plus intrathecal methotrexate on the same days in those patients treated for a relapse. The majority of patients (81%) had St Jude stage III or IV disease. Twenty patients (71%) achieved a complete clinical remission. Day 8 treatment was delayed in eight children and Day 15 treatment in five patients, both for a median of 7 days, mainly due to neutropenia. Ten patients relapsed after 42-311 days (median 105 days). Ten patients (36%) remained in remission for 353-712 days (median 487 days). Patients whose first relapse occurred after 6 months as well as those with limited disease had the best outcome. This simple 15-day chemotherapy schedule salvaged 36% of patients and significantly increased the overall cure rate of our Burkitt's lymphoma patients.

Spectrum and presentation of pediatric malignancies in the HIV era: experience from Blantyre, Malawi, 1998-2003.

BACKGROUND: Data on childhood cancers in Africa are sparse, particularly since the spread of HIV. We aimed to document the frequency of pediatric cancers presenting to a large central hospital in Malawi, detailing the presenting features, initial investigations, and HIV status of these children. PROCEDURE: A retrospective audit of the spectrum and clinical presentation of cancers among children (<16 years) seen at Queen Elizabeth's Central Hospital (QECH), between 1998 and 2003. RESULTS: Seven hundred seven children with cancer were seen, the number of cases per year increased over the time period; 50% (351) had Burkitt lymphoma, 13% (89) had retinoblastoma, and 9% (61) had Kaposi sarcoma, with a variety of other tumors comprising the remainder. Kaposi sarcoma markedly increased in frequency over time. Histological verification of diagnosis was available for 49% (348). The proportion of children with cancer who were tested for HIV increased over time, but varied by cancer type. Amongst those tested, the seroprevalence was 93% (52/56) for children with Kaposi sarcoma, 4% (11/289) for those with Burkitt lymphoma, 31% (8/26) for those with other non-Hodgkin lymphomas, 7% (1/15) for those with Hodgkin disease, and 5% (5/103) for those with other cancers. CONCLUSIONS: The number of cases seen per year has increased over the study period for almost all cancers, but in particular for Kaposi sarcoma. Burkitt lymphoma remains the commonest pediatric tumor in Malawi. In the case of Burkitt lymphoma, non-Hodgkin lymphoma, and Kaposi sarcoma there is a significant difference in the presentation of HIV-seropositive and -seronegative children.

Effect of antibiotic treatment of subclinical mastitis on human immunodeficiency virus type 1 RNA in human milk.

Although subclinical mastitis is associated with increased HIV load in breast milk, it is not known whether empirical treatment with antibiotics will reduce breast milk HIV load. We examined the effect of antibiotic treatment for subclinical mastitis on HIV load in breast milk. Seventy-five HIV-infected post-partum women in Malawi with subclinical mastitis were treated with oral amoxicillin/clavulanic acid and were followed between 1 and 24 weeks later. Breastmilk HIV-1 load and sodium concentration were measured and microbiological studies were performed at presentation. At 1 week (n = 34), the proportion of women with elevated breast milk leukocyte counts decreased significantly to 41.2% (p < 0.0001) and there was a nonsignificant increase in breast milk HIV-1 RNA load (p = 0.9264) and sodium concentration (p = 0.08) in the affected breast. At 4 to 12 weeks (n = 63), breast milk HIV-1 RNA load and sodium concentration decreased significantly (p < 0.05) and 17.5% had elevated breast milk leukocyte counts. Treatment with amoxicillin/clavulanic acid was associated with a significant decrease in inflammation of the breast, but breast milk HIV load remained elevated despite a significant decrease from baseline. These findings have important implications regarding how mothers should be counselled on safety of resuming breastfeeding after resolution of subclinical mastitis.

Epidemiology and microbiology of subclinical mastitis among HIV-infected women in Malawi.

The epidemiology and microbiology of subclinical mastitis, a risk factor for perinatal HIV transmission, have not been well characterized. In all, 250 HIV-infected women were followed from two weeks to 12 months postpartum in Blantyre, Malawi, and subclinical mastitis was assessed by breast milk leukocyte counts. The point prevalence of subclinical mastitis at 2, 4, 6, 10, and 14 weeks, and 6, 9, and 12 months was 12.2%, 7.8%, 6.8%, 3.7%, 10.6%, 5.1%, 4.9%, and 1.9%, respectively (P = 0.002), and 27.2% of women had at least one episode of subclinical mastitis. There was no significant relationship between maternal plasma HIV load or parity and subclinical mastitis. Staphylococcus aureus was isolated in 30% of women with subclinical mastitis, and the proportion of women with positive cultures decreased during follow-up (P = 0.02). Subclinical mastitis is prevalent among breastfeeding mothers and further studies are needed to characterize the differences between infectious and non-infectious subclinical mastitis.

Haematological profiles of the people of rural southern Malawi: an overview.

An integrative review of the results of two published and two unpublished studies of anaemia in children, adolescent females, pregnant women and adults living in southern Malawi is presented. Anaemia was universally present in all age-groups, with the higher prevalences in infants (100%) and adolescent primigravidae (93.8%). Nutritional deficits of iron and vitamin A were major contributory factors but chronic malarial haemolysis also significantly contributed to the anaemia. Among boys, anaemia was more common among those with glucose-6-phosphate-dehydrogenase (G6PD) deficiency than in those without this deficiency (P<0.002). This enzymopathy, which occurred in 23.5% [95% confidence interval (CI)=16.7%-30.1%] of the male and 30% (CI=17.3%-42.7%) of the female infants examined, was also associated with neonatal jaundice. The overall prevalences of the-alpha(3.7)/alphaalpha and -alpha(3.7)/-alpha(3.7) thalassaemia genotypes were estimated at 41.0% (CI=28.3%-53.7%) and 8.7% (CI=1.5%-15.9%), respectively. Haemoglobin AS was present in 18.1% (CI=12.8%-23.4%) of the infants and haemoglobin SS in 2.5% (CI=1.4%-3.6%). As the prevalence of infection with Plasmodium falciparum was significantly higher in infants with haemoglobin AS than in those with AA (21.4% v. 6.7%; P<0.001), an increased risk of early-onset moderate parasitaemias in young infants probably stimulates the development of immunity, protecting older heterozygotes from severe malarial infection. Innovative community approaches are required to break the cycle of ill health that anaemia supports in those living in rural areas of southern Malawi. Interventions in adolescent girls could be of particular importance, as they could break the cycle in both pregnant women and their infants.

CO2 production during acute infection in malnourished Malawian children.

OBJECTIVE: This study tested the hypotheses that the rate of CO2 production is less in marasmic children with acute infection when compared to well-nourished children, but greater when compared to uninfected marasmic children. DESIGN: A descriptive comparison of children aged 12-60 months who had their rates of CO2 production measured using a stable isotope tracer dilution method while receiving feedings. Body mass index (BMI) was the best measure of lean body mass available in this study. SETTING: Queen Elizabeth Central Hospital, Blantyre, Malawi. SUBJECTS: A total of 56 children were studied, 28 with marasmus and acute infection, 16 with marasmus, and 12 well nourished with acute infection. Those with acute infection had malaria, pneumonia, or sepsis. RESULTS: Well-nourished children with acute infection produced more CO2 than marasmic children (344+/-60 vs 225+/-65 mmol CO2/h, mean+/-s.d., P<0.001; 24.2+/-4.6 vs 18.4+/-5.4 mmol CO2/BMI h, P=0.001). However, the rate of CO2 production in marasmic children with acute infection was not greater than in uninfected marasmic children (225+/-65 vs 228+/-61 mmol CO2/h). The observed rate of CO2 production was greater than that which could be produced from the dietary intake alone (29.6 vs. 25.8 mmol CO2/kg h). CONCLUSIONS: Marasmic children do not increase energy expenditure in response to acute infection, as well-nourished children do. Dietary energy provided to marasmic children should be at least 420 kJ/kg day.

Risk factors for fetal anaemia in a malarious area of Malawi.

The prevalence of infants born with low cord haemoglobin (fetal anaemia) is high in areas where malaria and iron deficiency anaemia in pregnancy are common. The objective of the present study was to determine risk factors for fetal anaemia in an area of high malaria transmission in southern Malawi. A case control study was undertaken with fetal anaemia defined as cord haemoglobin (Hb) < 12.5 g/dl. Between March 1993 and July 1994, pregnant women attending the study hospitals for the first time in that pregnancy were enrolled. Data on socio-economic status, anthropometry, previous obstetric history and current pregnancy were collected. Malaria parasitaemia, Hb levels and iron status were measured in maternal blood at recruitment and delivery and in umbilical venous blood. Fetal anaemia occurred in 23.4% of babies. Mean (SD) cord Hb was 13.6 g/dl (1.83). Factors associated with fetal anaemia were: birth in the rainy season [adjusted odds ratio (AOR) 2.33, 95% CI 1.73-3.14], pre-term delivery (AOR 1.60, 1.03-2.49), infant Hb < 14 g/dl at 24 hours (AOR 2.35, 1.20-4.59), maternal Hb at delivery < 8 g/dl (AOR 1.61, 1.10-2.42) or <11 g/dl (AOR 1.60, 1.10-2.31). A higher prevalence of fetal anaemia occurred with increasing peripheral Plasmodium falciparum parasite density (p=0.03) and geometric mean placental parasite densities were higher in babies with fetal anaemia than in those without (3331 vs 2152 parasites/microl, p=0.07). Interventions should aim to reduce fetal anaemia by improving malaria and anaemia control in pregnancy and by addressing the determinants of pre-term delivery.

Teaching tuberculosis control to medical undergraduates: the Malawi experience.

SETTING: National Tuberculosis (TB) Control Programme (NTP) and College of Medicine (COM), Malawi. OBJECTIVES: To develop a TB/HIV module, incorporating TB control and the DOTS strategy, for 4th year medical undergraduates. To describe 1) the way in which the module was developed, 2) the contents and structure of the module, 3) the experience of teaching the module from 2000-2002, and d) the financial costs to the NTP. DESIGN: A descriptive study. RESULTS: The TB/HIV module, including the teaching manual, resource materials and undergraduate assessments, was developed between June and December 1999 by NTP, College of Medicine, interested stakeholders and an external consultant. The module was well received by medical undergraduates. Student knowledge, based on pre-module and post-module assessments, increased to satisfactory levels. Novel aspects of teaching, which included reading chapters in class followed by student-led knowledge reviews, modular assessments and using NTP staff as facilitators, were highly rated. The cost of developing the module was 14,070 US dollars, and the recurrent annual cost of teaching the module was 900 US dollars. CONCLUSION: The results show that a national tuberculosis control programme can work effectively with an academic medical institution in teaching medical undergraduates the important principles of country-wide TB control.

Epidemiology of measles in Blantyre, Malawi: analyses of passive surveillance data from 1996 to 1998.

Measles surveillance data in Blantyre, Malawi were reviewed for 1996-8 to describe the epidemiology of infection and to estimate vaccine efficacy (VE) by the screening method. A total of 674 measles cases were reported to the Blantyre District Health Office during this period. Age distribution showed that 108 (16.1%) of the cases were aged less than 1 year. The median age was 5 years. Eighty percent of the cases between 1 and 19 years had been previously vaccinated. VE was 68.6% (95% CI, 527-792) for children 12-23 months of age and 67.3% (95% CI, 48.3-79.3) for infants 9-11 months of age. Reasons for this low vaccine efficacy are discussed. Previous vaccination history was negatively associated with the risk for developing cough during measles infection (odds ratio (OR), 0.30; 95% CI, 0.09-0.91), diarrhoea (OR, 0.64; CI, 0.44-0.95) and pneumonia (OR, 0.40; CI, 0.25-0.62). Logistic regression analysis showed that pneumonia in adults was negatively associated with vaccination history. The passive surveillance system for measles in Malawi was useful to describe the epidemiology of measles.

Childhood tuberculosis in Malawi: nationwide case-finding and treatment outcomes.

SETTING: All 43 non-private hospitals (three central, 22 [corrected] district and 18 [corrected] mission) in Malawi that register and treat adult and paediatric TB cases. OBJECTIVE: To assess the rate, pattern and treatment outcome of childhood TB case notifications in Malawi in 1998. DESIGN: Retrospective data collection using TB registers, treatment cards and information from health centre registers. Information was collected on number of cases, types of TB and treatment outcomes using standardised definitions. RESULTS: There were 22,982 cases of TB registered in Malawi in 1998, of whom 2,739 (11.9%) were children. Children accounted for 1.3% of all case notifications with smear-positive pulmonary TB (PTB), 21.3% with smear-negative PTB and 15.9% with extra-pulmonary TB (EPTB). Estimated rates of TB in children were 78/ 100,000 in those aged less than one year, 83/100,000 in those aged 1-4 years and 33/100,000 in those aged 5-14 years. A significantly higher proportion of TB cases was diagnosed in central hospitals. Only 45% of children completed treatment. There were high rates of death (17%), default (13%) and unknown treatment outcomes (21%). Treatment outcomes were worse in younger children and in children with smear-negative PTB. Treatment completion was best (76%) and death rates lowest (11%) for the 127 children with smear-positive PTB. CONCLUSION: Childhood TB is common in Malawi and treatment outcomes are poor. Research should be directed towards improved diagnosis and follow-up of children with TB, and the National TB Programme should support appropriate management of childhood contacts of smear positive PTB cases.

Changes in haemoglobin levels in infants in Malawi: effect of low birth weight and fetal anaemia.

OBJECTIVES: To examine the effect of low birth weight (LBW) and fetal anaemia (FA) on haemoglobin (Hb) patterns in infancy. To study the additional contribution of other risk factors known at birth. To examine the effect of iron supplementation during infancy on Hb levels. METHODS: A stratified cohort of infants in Malawi (83 with LBW (< 2500 g), 111 with FA (cord Hb < 125 g/l), 31 with both LBW and FA, and 176 controls) was followed during infancy. Hb levels were measured at about 2, 4, 6, 9, and 12 months of age. Repeated measures models were used to describe the changes in Hb levels over time. RESULTS: The mean Hb concentration in the control group was 95.5 g/l (95% confidence interval (CI) 92.5 to 98.5) at 2 months, 86.9 g/l (95% CI 84.4 to 89.4) at 9 months, and 898 g/l (95% CI 874 to 92.2) at 12 months. Differences between LBW infants and controls increased over time (difference at 12 months: 5.5 g/l (95% CI 1.3 to 9.7)). Infants with FA had borderline significantly lower Hb at 2 months (p = 0.07), but at 6 months their levels were similar to those of controls. The LBW infants and those with FA had the lowest Hb levels (difference from controls at 12 months 7.9 g/l). Parity, placental and maternal malaria at delivery, and sex significantly affected Hb levels after adjustment for LBW and FA. After iron supplementation, Hb significantly increased. CONCLUSIONS: Antimalarial control and iron supplementation throughout pregnancy should be increased to reduce the incidence of infant anaemia and improve child development and survival.

Detection and characterization of rotaviruses in hospitalized neonates in Blantyre, Malawi.

In five separate fecal collections spanning three years, group A rotaviruses were detected by enzyme-linked immunosorbent assay in 35 (25%) of 142 specimens obtained from nondiarrheic, hospitalized neonates in Blantyre, Malawi. Molecular characterization of each strain identified, for the first time in neonates, a short electropherotype, genotype P[6], G8 strain type, similar to the dominant, cocirculating community strain detected in symptomatic infants in Blantyre. Partial sequence analysis of the VP4 and NSP4 genes of neonatal and community strains failed to identify changes which could explain the differences in clinical outcome. Neonatal serotype G8 rotaviruses should be considered as potential rotavirus vaccine candidates for use in Malawi.

Detection of group C rotavirus in children with acute gastroenteritis in Blantyre, Malawi.

Among 606 children who were treated for acute gastroenteritis at the Queen Elizabeth Central Hospital in Blantyre, Malawi, Group C rotavirus (Gp C RV) was detected by enzyme-linked immunosorbent assay in fecal specimens from 16 (3.9%) of 408 inpatients and in 4 (2.0%) of 198 outpatients. Thirteen (65%) children excreting Gp C RV were coinfected with Group A rotavirus.

Plasma HIV burden in Malawian children co-infected with rotavirus.

Fifty-eight HIV-infected children with acute rotavirus diarrhea were tested for plasma HIV RNA. There was no difference between acute and convalescent mean viral loads, and little change in CD4 cell counts. Compared with the 16 children who died within 4 weeks, 31 survivors had slightly lower viral loads at presentation and significantly higher CD4 cell counts. Low CD4 cell counts, but not HIV-1-RNA concentrations, were predictive of Death. Local, enteric rotavirus infection did not appear to affect blood HIV viral load or CD4 cell counts in this small group of children.

Impact of vitamin A supplementation on anaemia and plasma erythropoietin concentrations in pregnant women: a controlled clinical trial.

INTRODUCTION: Although studies suggest that vitamin A or its metabolites influence the synthesis of erythropoietin in vitro and in animal models, it is unclear whether vitamin A supplementation increases plasma erythropoietin concentrations in humans. OBJECTIVE: To determine whether daily vitamin A supplementation increases plasma erythropoietin concentrations in pregnant women with a high prevalence of anaemia. METHODS: A randomized, double-blind, controlled clinical trial was conducted to examine the effect of daily vitamin A (3000 microg retinol equivalent), iron (30 mg), and folate (400 microg) versus iron (30 mg) and folate (400 microg) (control) on haemoglobin and plasma erythropoietin concentrations in 203 pregnant women in Malawi, Africa. RESULTS: Mean gestational age at enrollment was 23 wk, at which time 50% of the women were anaemic (haemoglobin <110 g/L). Mean (+/-SEM) change in haemoglobin from enrollment to 38 wk was 4.7+/-1.6 g/L (p=0.003) and 7.3+/-2.3 g/L (p=0.003) in the vitamin A and control groups, respectively. Mean change in plasma erythropoietin concentrations from enrollment to 38 wk was 2.39+/-5.00 (p=0.63) and -2.87+/-3.92 IU/L (p=0.46) in the vitamin A and controls groups, respectively. There were no significant differences between vitamin A and control groups in the slope of the regression line between log10 erythropoietin and haemoglobin at enrollment or 38 wk, and between enrollment and follow-up within either group. CONCLUSIONS: Vitamin A supplementation does not appear to increase haemoglobin and plasma erythropoietin concentrations among pregnant women with a high prevalence of anaemia in Malawi.

Effect of concomitant HIV infection on presentation and outcome of rotavirus gastroenteritis in Malawian children.

BACKGROUND: Rotaviruses represent important causes of severe diarrhoea in early childhood. We examined the effect of HIV infection on the presentation and outcome of rotavirus gastroenteritis in Malawian children. METHODS: Children younger than 5 years who were treated for acute gastroenteritis at the Queen Elizabeth Central Hospital in Blantyre from July, 1997, to June, 1999, were enrolled. Children with rotavirus diarrhoea, with and without HIV infection, were followed up for up to 4 weeks after hospital discharge. Rotavirus disease severity (assessed with a 20-point score), duration of rotavirus shedding, and seroresponse to rotavirus were compared between HIV-infected and HIV-uninfected children. FINDINGS: 786 inpatients (median age 8 months, 271 [34%] of whom were HIV-1-infected) and 400 outpatients (median age 9 months, 65 [16%] of whom were HIV-infected) were enrolled. Rotavirus was detected less frequently among HIV-infected children (102 of 336 [30%]) than among HIV-uninfected children (348 of 850 [41%], (relative risk 0.71 [95% CI 0.53-0.87], p=0.0007). There were no differences in rotavirus disease severity for hospitalised children with and without HIV infection, but HIV-infected children were more likely to die during follow-up (11/50 [22%]) than HIV-uninfected children (0/61, p<0.0001). Of 29 HIV-infected and 45 HIV-uninfected children who completed follow-up, six (21%) HIV-infected children shed rotavirus, compared with two (4%) HIV-uninfected children (4.66 [1.01-21.51], p=0.05), but shedding was not associated with diarrhoea. Three-quarters of children exhibited a four-fold rise of serum IgG or IgA to rotavirus, which did not vary by HIV status. INTERPRETATION: Malawian children with concomitant HIV infection resolved acute rotavirus infections. Rotavirus vaccine safety and immunogenicity in HIV-infected infants should now be determined.

An analysis of intra-uterine growth retardation in rural Malawi.

OBJECTIVE: (1) To describe the sex-specific, birth weight distribution by gestational age of babies born in a malaria endemic, rural area with high maternal HIV prevalence; (2) to assess the contribution of maternal health, nutritional status and obstetric history on intra-uterine growth retardation (IUGR) and prematurity. METHODS: Information was collected on all women attending antenatal services in two hospitals in Chikwawa District, Malawi, and at delivery if at the hospital facilities. Newborns were weighed and gestational age was assessed through post-natal examination (modified Ballard). Sex-specific growth curves were calculated using the LMS method and compared with international reference curves. RESULTS: A total of 1423 live-born singleton babies were enrolled; 14.9% had a birth weight <2500 g, 17.3% were premature (<37 weeks) and 20.3% had IUGR. A fall-off in Malawian growth percentile values occurred between 34 and 37 weeks gestation. Significantly associated with increased IUGR risk were primiparity relative risk (RR) 1.9; 95% CI 1.4--2.6), short maternal stature (RR 1.6; 95% CI 1.0--2.4), anaemia (Hb<8 g/dl) at first antenatal visit (RR 1.6; 95% CI 1.2--2.2) and malaria at delivery (RR 1.4; 95% CI 1.0--1.9). Prematurity risk was associated with primiparity (RR 1.7; 95% CI 1.3--2.4), number of antenatal visits (RR 2.2; 95% CI 1.6--2.9) and arm circumference <23 cm (RR 1.9; 95% CI 1.4--2.5). HIV infection was not associated with IUGR or prematurity. CONCLUSION: The birth-weight-for-gestational-age, sex-specific growth curves should facilitate improved growth monitoring of newborns in African areas where low birth weight and IUGR are common. The prevention of IUGR requires improved malaria control, possibly until late in pregnancy, and reduction of anaemia.

The Australian contribution towards medical training in Malawi.

Malawi opened its only medical school in April 1991, despite opposition from some who believed that the venture was a waste of precious resources. From 1986, selected Malawi students had done preclinical training in Australia, the United Kingdom and South Africa, returning to Malawi for the clinical phase of their training. From 1994, Malawi students have been able to do the whole undergraduate medical course in their own country. Twenty Malawi medical students attended the medical schools of the University of Adelaide and Flinders University in South Australia between 1991 and 1996. All but three of those students are now qualified doctors working in various capacities in healthcare delivery service in Malawi.