Amelia Catharine Crampin

Serological data provide important insights into SARS-CoV-2 transmission and immunity, particularly in regions with limited routine surveillance such as sub-Saharan Africa. However, antibody waning and boosting following reinfection or vaccination remain poorly characterised, complicating interpretation of serological measurements. Improved understanding of these dynamics is critical for accurate epidemiological inference. Modelling longitudinal serological data provides a means to quantify antibody kinetics and reconstruct infection histories. We analysed 15,679 neutralising antibody (nAb) titres from 1,675 unvaccinated, HIV-uninfected participants in urban (Lilongwe) and rural (Karonga) Malawi (February 2021–April 2022). NAb titres against ancestral B.1, Beta, Delta, and Omicron (BA.1/BA.2) viruses were measured using an HIV-based SARS-CoV-2 pseudotyped virus neutralisation assay. A multi-level Bayesian model was used to reconstruct infection histories and antibody kinetics. The model identified 429 infections (95% credible interval 417–441), including 39 (9.1%) that had not been identified by traditional seroconversion-based thresholds. Antibody levels waned rapidly, with 48% (0.403–0.560) of the acute boost remaining after three months and only 5% (0.027–0.098) after one year. Pre-Omicron infections generated stronger antibody boosts than Omicron infections. Responses varied, with individuals clustering into low and high responders. Cross-reactive responses extended across substantial antigenic distances - Omicron infections induced broader immunity. Seroincidence was higher in Lilongwe than in Karonga (0.41 vs. 0.27 infections per person per three months), driven by the early 2022 Omicron wave. Reinfections were common, particularly among adults and urban residents. SARS-CoV-2 nAb responses following infection were heterogeneous and declined rapidly. This rapid waning underscores the importance of vaccination for sustained protection, while cross-reactivity suggests only partial immunity from prior variants. Identifying reinfections is essential for understanding transmission and finding populations at higher repeat infection risk, particularly where routine surveillance is limited.

Background: Emerging evidence from African and other low-and-middle -income populations suggests that a substantial proportion of apparent type 2 diabetes (T2D) cases occur at lower body mass index (BMI) (< 25kg/m2) and younger ages (<40 years) than seen in high-income populations. Whether this represents predominance of a distinct aetiology or reflects underlying population structure differences is unknown. We determined whether there is a disproportionate burden of T2D in non-overweight and/or younger individuals in an African population compared with a high-income population.Methods: We analysed data from 24,985 adults aged ≥18 years from a Malawian population-based cross-sectional study (MEIRU) conducted between 2013 and 2016, and 41,937 adults from the National Health and Nutrition Examination Survey (NHANES) conducted between 2009 and 2023. We defined T2D as a fasting blood glucose level of ≥ 7mmol/L and/or a previously reported diagnosis. We computed the crude and age-standardised prevalences across BMI groups and used logistic regression to compute adjusted odds ratios (95% CI) for the associations between participant characteristics and diabetes.Results: A higher proportion of T2D cases in Malawi occurred in non-overweight individuals (34% [95% CI: 30,38%]) compared with Black (12% [8·7,18%]) and non-Black Americans (11% [9·3,13%]), which mirrored the underlying population distribution (Malawi non-overweight 69% [68·2,69·4%]) vs Black 23% [21·6,25·1%]) and non-Black Americans 30% [28·5,31·4%]). A greater proportion of T2D cases in Malawi occurred in individuals aged <40 years (24% [21,28%]) compared with Black (12% [8·2,17%]) and non-Black Americans (8% [6·8,9·6%]), also reflecting population structure, (Malawi aged <40y: 69% [69·1,70·2%]), compared to Black Americans 42% [39·9,44·5%]) and non-Black Americans 37% [35·6-38·5%]).Age-standardised prevalence of T2D was lower among non-overweight Malawians than in the US population. For BMI <18·5 kg/m², a suggested cut-off for malnutrition-related diabetes, prevalence in Malawians was 1·3% (95% CI: 0·7, 2.0%) compared to 4·1% (2·2, 7·7%) in Black and 3·7% (1·8,7·6%) in non-Black Americans. Similarly, for BMI 18·5- 24·9 kg/m2, Malawian prevalence (2%) remained lower than that of Black and non-Black Americans (9·2% and 4·1%, respectively).Higher BMI and older age were strongly associated with T2D across all populations. Each 1 SD increase in BMI was associated with increased odds of T2D (OR 95% CI: 1·9, [1·8-2·0], 1·6, [1·4-1·8], and 1·9, [1·8-2·0]) as was each 1SD increase in age (2·1,[1·9-2·3]; 2·8, [2·4-3·3] and 2·9,[2·7-3·1]) in Malawian, Black Americans and non-Black Americans respectively.Conclusion: The higher proportion of T2D cases occurring in leaner and younger individuals in African compared to high-income populations reflects differences in population BMI and age distribution rather than increased individual susceptibility at lower BMIs or younger ages.

Background: Serological data provide important insights into SARS-CoV-2 transmission and immunity, particularly in regions with limited routine surveillance such as sub-Saharan Africa. However, antibody waning and boosting following reinfection or vaccination remain poorly characterised, complicating interpretation of serological measurements. Improved understanding of these dynamics is critical for accurate epidemiological inference. Modelling longitudinal serological data provides a means to quantify antibody kinetics and reconstruct infection histories. Methods: We analysed 15,679 neutralising antibody (nAb) titres from 1,675 unvaccinated, HIV-uninfected participants in urban (Lilongwe) and rural (Karonga) Malawi (February 2021 - April 2022). NAb titres against ancestral B.1, Beta, Delta, and Omicron (BA.1/BA.2) viruses were measured using an HIV-based SARS-CoV-2 pseudotyped virus neutralisation assay. A multi-level Bayesian model was used to reconstruct infection histories and antibody kinetics. Findings: The model identified 429 infections (95% credible interval 417-441), including 39 (9·1%) that had not been identified by traditional seroconversion-based thresholds. Antibody levels waned rapidly, with 48% (0·403-0·560) of the acute boost remaining after three months and only 5% (0·027-0·098) after one year. Pre-Omicron infections generated stronger antibody boosts than Omicron infections. Responses varied, with individuals clustering into low and high responders. Cross-reactive responses extended across substantial antigenic distances - Omicron infections induced broader immunity. Seroincidence was higher in Lilongwe than in Karonga (0·41 vs. 0·27 infections per person per three months), driven by the early 2022 Omicron wave. Reinfections were common, particularly among adults and urban residents. Interpretation: SARS-CoV-2 nAb responses following infection were heterogeneous and declined rapidly. This rapid waning underscores the importance of vaccination for sustained protection, while cross-reactivity suggests only partial immunity from prior variants. Identifying reinfections is essential for understanding transmission and finding populations at higher repeat infection risk, particularly where routine surveillance is limited.

Background: Verbal autopsy, where a close caregiver or relative of someone who recently died reports on the signs, symptoms and circumstances preceding death, is useful for producing population‐based cause of death estimates. However, the performance of verbal autopsy for older adult deaths is poorly understood.Objectives: To evaluate the performance of verbal autopsy in assigning cause of death for adults aged 50+ in a rural area of Malawi.Methods: Cause of death was assigned to each death with a verbal autopsy in the Karonga Health and Demographic Surveillance site between 2002 and 2017 using two methods: (1) Physician review and (2) in silico verbal autopsy (a Bayesian probabilistic model). We calculated uncertainty in cause of death assignment for each method and calculated disagreement in cause of death between methods. Analyses were stratified by age group and sex.Results: A total of 2378 adult deaths were included (1360 aged 50+). Cause of death assignment showed greater uncertainty at older ages in both methods. For example, 59.7% of men aged 80+ were assigned a specific cause of death using physician review, versus 77.5% of men aged 30–49. Population‐level, broad cause of death distributions were similar across methods, but at the individual level there was over 30% disagreement on broad cause of death categories in those aged 50+.Conclusions: Verbal autopsy becomes more uncertain at assigning cause of death at older ages. The inclusion of any reports of medical diagnoses from a doctor and using a two‐stage process of cause of death assignment (with simple cause of deaths assigned using algorithms and more complex cases being reviewed by physicians) could improve cause of death ascertainment using verbal autopsy at older ages.

: Supplementary methods of mortality surveillance are needed in settings with incomplete death registration. Local perceptions of mortality levels might be useful indicators of excess deaths. Early in the COVID-19 pandemic, we developed a survey question asking respondents to evaluate the recent frequency of funerals in their community relative to a pre-pandemic baseline. We asked this question of more than 400 residents of Karonga district in Malawi, who were interviewed up to five times during a panel survey conducted by mobile phone between June 2020 and May 2021. The proportion of respondents reporting more funerals than usual in their community increased in early 2021, indicating excess mortality not otherwise detected by case-based surveillance and only partially visible in reports of inpatient deaths at health facilities. Systems assessing changes in perceptions of mortality can be rapidly established during an epidemic and may help detect excess deaths at local levels.

Volume 51 - Article 14 | Pages 459–500

Here we describe a dataset from a cross-sectional survey carried out by the Malawi Epidemiology and Intervention Research Unit, to describe the burden and determinants of hypertension, diabetes and lipid disorders in rural and urban Malawi. Data were collected between May 2013 and April 2017 through face-to-face data capture in participant’s homes in the Karonga district in northern Malawi (n=13,904) and Area 25 in the capital Lilongwe in the central region (n=16,660). Data collected included sections on demographic data, marital status and spouse information, household socio-economic status, smoking, indoor smoke exposure, alcohol use, diets, physical activity, pregnancy, medical history and family medical history. All blood samples were tested for glucose, lipids, electrolytes and full blood count. Data are available to researchers on reasonable request.

Proximity to family, household composition, and structure are often studied as outcomes and as explanatory factors in a wide range of scientific disciplines. Here, we describe a large longitudinal dataset (currently including data from over 70,000 individuals from 2004 to 2017), including data on household structure, proximity to kin, population density, and other socio-demographic factors derived from data from the Karonga Health and Demographic Surveillance Site (HDSS) in Northern Malawi. We present how the dataset is generated, list some examples of how it can be used, and provide information on the limitations that affect the types of analyses that can be carried out.

Background In sub-Saharan Africa, migration of young people is common and occurs for a variety of reasons. Research focus is often on international or long-distance internal migration; however, shorter moves also affect people’s lives and can reveal important information about cultures and societies. In rural sub-Saharan Africa, migration may be influenced by cultural norms and family considerations: these may be changing due to demographic shifts, urbanisation, and increased media access. Methods We used longitudinal data from a Health and Demographic Surveillance Site in rural northern Malawi to present a detailed investigation of migration in young people between 2004–2017. Our focus is on the cultural effects of gender and family, and separate migrations into short and long distance, and independent and accompanied, as these different move types are likely to represent very different events in a young person’s life. We use descriptive analyses multi-level multinomial logistic regression modelling. Results & conclusions We found two key periods of mobility 1) in very young childhood and 2) in adolescence/young adulthood. In this traditionally patrilocal area, we found that young women move longer distances to live with their spouse, and also were more likely to return home after a marriage ends, rather than remain living independently. Young people living close to relatives tend to have lower chances of moving, and despite the local patrilineal customs, we found evidence of the importance of the maternal family. Female and male children may be treated differently from as young as age 4, with girls more likely to migrate long distances independently, and more likely to accompany their mothers in other moves.