Prevalence and Associated Factors of Chronic Obstructive Pulmonary Disease Among Adults in Neno District, Malawi: A Cross-Sectional Analytical Study.

Introduction: Chronic obstructive pulmonary disease (COPD) continues to pose a global public health challenge. However, literature is scarce on the burden of COPD in Malawi. We assessed the prevalence and risk factors for COPD among adults in Neno, Malawi. Methodology: We conducted a population-based analytical cross-sectional study in Neno District between December 2021 and November 2022. Using a multi-stage sampling technique, we included 525 adults aged≥40 years. All participants underwent spirometry according to the American Thoracic Society (ATS) guidelines and were interviewed using the IMPALA questionnaire. For this study, we utilized the definition of COPD as a post-bronchodilator FEV1/FVC <0.70. We collected data using Kobo collect, exported to Microsoft Excel, and analysed using R software. We used descriptive statistics and logistic regression analysis; a p-value of <0.05 was considered statistically significant. Results: Out of 525 participants, 510 participants were included in the final analysis. Fifty-eight percent of the participants were females (n=296), and 62.2% (n=317) were between 40 and 49 years with a median (IQR) age of 46 (40-86). For patient characteristics, 15.1% (n=77) were current smokers, and 4.1% (n=21) had a history of pulmonary tuberculosis (PTB). Cough was the most commonly reported respiratory symptom (n=249, 48.8%). The prevalence of COPD was 10.0% (n=51) and higher (15.0%) among males compared to females (6.4%). Factors significantly associated with COPD were age 60 years and above (adjusted odds ratio [aOR] = 3.27, 95% CI: 1.48-7.34, p<0.004), ever smoked (aOR = 6.17, 95% CI:1.89-18.7, p<0.002), current smoker (aOR = 17.6, 95% CI: 8.47-38.4, p<0.001), and previous PTB (aOR = 4.42, 95% CI: 1.16-15.5, p<0.023). Conclusion: The cross-sectional prevalence of COPD in rural Malawi is high, especially among males. Factors significantly associated were older age (60 years and above), cigarette smoking, and previous PTB. Longitudinal studies are needed to better understand disease etiology and progression in this setting.

Global implementation research capacity building to address cardiovascular disease: An assessment of efforts in eight countries.

Cardiovascular diseases are the leading causes of morbidity and mortality worldwide, but implementation of evidence-based interventions for risk factors such as hypertension is lacking, particularly in low and middle income countries (LMICs). Building implementation research capacity in LMICs is required to overcome this gap. Members of the Global Research on Implementation and Translation Science (GRIT) Consortium have been collaborating in recent years to establish a research and training infrastructure in dissemination and implementation to improve hypertension care. GRIT includes projects in Ghana, Guatemala, India, Kenya, Malawi, Nepal, Rwanda, and Vietnam. We collected data from each site on capacity building activities using the Potter and Brough (2004) model, mapping formal and informal activities to develop (a) structures, systems and roles, (b) staff and infrastructure, (c) skills, and (d) tools. We captured information about sites' needs assessments and metrics plus program adaptations due to the COVID-19 pandemic. All sites reported capacity building activities in each layer of the Capacity Pyramid, with the largest number of activities in the Skills and Tools categories, the more technical and easier to implement categories. All sites included formal and informal training to build Skills. All sites included a baseline needs assessment to guide capacity building activities or assess context and inform intervention design. Sites implementing evidence-based hypertension interventions used common implementation science frameworks to evaluate implementation outcomes. Although the COVID-19 pandemic affected timelines and in-person events, all projects were able to pivot and carry out planned activities. Although variability in the activities and methods used existed, GRIT programs used needs assessments to guide locally appropriate design and implementation of capacity building activities. COVID-19 related changes were necessary, but strong collaborations and relationships with health ministries were maintained. The GRIT Consortium is a model for planning capacity building in LMICs.

Decentralized Heart Failure Management in Neno, Malawi.

Background: Cardiovascular disease (CVD) is a major cause of death in Malawi. In rural districts, heart failure (HF) care is limited and provided by non-physicians. The causes and patient outcomes of HF in rural Africa are largely unknown. In our study, non-physician providers performed focused cardiac ultrasound (FOCUS) for HF diagnosis and longitudinal clinical follow-up in Neno, Malawi. Objectives: We described the clinical characteristics, HF categories, and outcomes of patients presenting with HF in chronic care clinics in Neno, Malawi. Methods: Between November 2018 and March 2021, non-physician providers performed FOCUS for diagnosis and longitudinal follow-up in an outpatient chronic disease clinic in rural Malawi. A retrospective chart review was performed for HF diagnostic categories, change in clinical status between enrollment and follow-up, and clinical outcomes. For study purposes, cardiologists reviewed all available ultrasound images. Results: There were 178 patients with HF, a median age of 67 years (IQR 44 - 75), and 103 (58%) women. During the study period, patients were enrolled for a mean of 11.5 months (IQR 5.1-16.5), after which 139 (78%) were alive and in care. The most common diagnostic categories by cardiac ultrasound were hypertensive heart disease (36%), cardiomyopathy (26%), and rheumatic, valvular or congenital heart disease (12.3%).At follow-up, the proportion of New York Heart Association (NYHA) class I patients increased from 24% to 50% (p < 0.001; 95% CI: 31.5 - 16.4), and symptoms of orthopnea, edema, fatigue, hypervolemia, and bibasilar crackles all decreased (p < 0.05). Conclusion: Hypertensive heart disease and cardiomyopathy are the predominant causes of HF in this elderly cohort in rural Malawi. Trained non-physician providers can successfully manage HF to improve symptoms and clinical outcomes in limited resource areas. Similar care models could improve healthcare access in other rural African settings.

Bivariate logistic regression model diagnostics applied to analysis of outlier cancer patients with comorbid diabetes and hypertension in Malawi.

The joint occurrence of diabetes and hypertension conditions in a patient is common. The two diseases share a number of risk factors, and are hence usually modelled concurrently using bivariate logistic regression. However, the postestimation assessment for the model, such as analysis of outlier observations, is seldom carried out. In this article, we apply outlier detection methods for multivariate data models to study characteristics of cancer patients with joint outlying diabetes and hypertension outcomes observed from among 398 randomly selected cancer patients at Queen Elizabeth and Kamuzu Central Hospitals in Malawi. We used R software version 4.2.2 to perform the analyses and STATA version 12 for data cleaning. The results showed that one patient was an outlier to the bivariate diabetes and hypertension logit model. The patient had both diabetes and hypertension and was based in rural area of the study population, where it was observed that comorbidity of the two diseases was uncommon. We recommend thorough analysis of outlier patients to comorbid diabetes and hypertension before rolling out interventions for managing the two diseases in cancer patients to avoid misaligned interventions. Future research could perform the applied diagnostic assessments for the bivariate logit model on a wider and larger dataset of the two diseases.

Perceived readiness for diabetes and cardiovascular care delivery in Mangochi, Malawi: multicentre study from healthcare providers' perspectives.

BACKGROUND: Despite the expected prevalence rise of 98.1% for diabetes between 2010 and 2030 in sub-Saharan Africa (SSA) and the anticipated rise of both diabetes and cardiovascular diseases (CVDs) in Malawi from their current figures ( 5.6 and 8.9%; respectively), data on the readiness of health facilities to provide diabetes and cardiovascular diseases in Mangochi district is not available. Therefore, this study aimed to assess the readiness of health facilities to provide services for diabetes and cardiovascular diseases. METHODS: An exploratory study was conducted from July to early September 2021 in 34 health facilities in Mangochi, Malawi. Forty-two participants were purposefully selected. They included medical officers, clinical officers, medical assistants, and registered nurses. The study used semi-structured interviews (for qualitative data) with a checklist (for quantitative data) to provide information about the readiness of services (such as guidelines and trained staff, drugs, diagnosis capacity and equipment, essential medicines, community services, and education/counseling).The thematic content analysis and basic descriptive statistics were carried out. RESULTS: The following main theme emerged from the qualitative part: low use of diabetes-cardiovascular disease (CVD) services. This was due to: health facility factors (shortage of drugs and supplies, poor knowledge, few numbers and lack of training of providers, and absent copies of guidelines), patients factors (poor health-seeking behaviour, lack of education and counseling for many), and community factors (very limited community services for diabetes and CVDs, lack of transport policy and high transportation costs). Data from the checklists revealed low readiness scores across domains (below the 75% target) in diabetes and cardiovascular diseases: trained staff and guidelines (26.5% vs. 32.4%); diagnosis capacity and equipment (63.7% vs. 66.2%); essential medicines (33.5% vs. 41.9%), and community services, and education and counseling (37.5% vs. 42.5%). CONCLUSION: There were several noticeable shortfalls identified in the readiness of health facilities to provide diabetes and cardiovascular disease services in Mangochi health facilities. Any future intervention in diabetes-cardiovascular disease care in these areas must include these elements in its basic package.

Training in eight low-and middle-income countries: lessons learned from a pilot study using the WHO-TDR dissemination and implementation massive open online course.

Introduction: Non-communicable diseases (NCDs) are a leading cause of morbidity and mortality in low-and middle- income countries (LMICs). Despite this, a lack of funding, training and mentorship for NCD investigators in LMICs exists. In an effort to gain knowledge and skills to address these gaps, participants from the Global Research on Implementation and Translation Science (GRIT), a consortium of studies in eight LMICs and their networks, attended the dissemination and implementation (D&I) massive open online course (MOOC) developed by the Special Programme for Research and Training in Tropical Diseases at the World Health Organization to strengthen D&I capacity building. Here, we report on the pilot of this MOOC, which was implemented during the SARS COVID-19 pandemic from April- November 2020. Methods: Participants completed pre-and post-training questionnaires to assess self-reported D&I competencies, general research skills, and research mentor access and quality. D&I competencies were measured by use of a scale developed for a US-based training program, with change in competency scores assessed by paired t test. We used univariate statistics to analyze the data for all other outcomes. Results: Of the 247 participants enrolled, 32 (13%) completed all course requirements, 21 (9%) completed the pre-and post-surveys and are included in the analysis. D&I competency scores suggest improvement for those who had complete pre- and post-assessments. Trainee's average score on the full competency scale improved 1.45 points (0-5 scale) from pre- to post-test; all four subscales also showed evidence of improvements. There were small but not significant increases in competencies for grant writing, proposal/ manuscript writing and presentations from pre- to post-test assessment. 40% of trainees reported access to a research mentor and 12% reported access to a D&I specific mentor. Participants reported barriers (e.g., unstable internet access and challenges due to COVID-19) and facilitators (e.g., topical interests, collaboration with colleagues) to completing the MOOC. Conclusions: Although COVID-19 affected program usage and completion, the MOOC was feasible. We also had signals of effectiveness, meaning among LMIC participants completing the course, there was improvement in self-report D&I competency scores. Recommendations for future D&I trainings in LMICs include (1) adding more topic specific modules (i.e., NCD research, general research skills) for scalability; (2) fostering more collaboration with participants across LMICs; and (3) establishing partnerships with D&I mentors for course participants.

High prevalence of dyslipidaemia among persons with diabetes mellitus and hypertension at a tertiary hospital in Blantyre, Malawi.

BACKGROUND: Dyslipidaemia drives the process of atherosclerosis, and hence a significant modifiable risk factor complicating hypertension and diabetes. In Malawi, the prevalence, screening and management of dyslipidaemia among persons with diabetes mellitus have not been reported. This study aimed to investigate the prevalence, biochemical characteristics, screening and management practices for dyslipidaemia among persons with diabetes mellitus, hypertension, and diabetes mellitus and hypertension comorbidity at Queen Elizabeth Central hospital in Blantyre, Malawi. METHODS: This was a cross-sectional study conducted in 2021. A total of 256 adult participants (diabetes mellitus = 100); hypertension = 100; both conditions = 56) were included. Medical data and anthropometric measurements were recorded. Blood samples were analysed for HbA1C and serum lipids. Associated risk factors for dyslipidaemia were also assessed. RESULTS: Dyslipidaemia was prevalent in 58%, 55%, and 70% of participants with diabetes mellitus, hypertension, and both conditions. Low-density lipoprotein cholesterol (LDL-C) dyslipidaemia was the most common in all participant groups. Participants with both diabetes and hypertension had 2.4 times (95% CI 1.2-4.6) increased risk of LDL-C dyslipidaemia than those with diabetes alone (p < 0.02). Being overweight or obese and age over 30 years were risk factors for dyslipidaemia in participants with diabetes mellitus alone (OR 1.3 (95% CI 1.1-1.6), p < 0.04, and OR 2.2 (95% CI 1.2-4.7) (p < 0.01), respectively. Overweight and obesity predicted LDL-C dyslipidaemia in hypertensive patients (OR 3.5 (95% CI 1.2-9.9) p < 0.001). Poorly controlled hypertension and the use of beta-blockers and thiazide diuretics predicted dyslipidaemia among patients with both diabetes mellitus and hypertension (OR 6.50 CI 1.45-29.19; and OR 5.20 CI 1.16-23.36 respectively). None of the participants had a lipogram performed before the study or were on lipid-lowering therapy. CONCLUSIONS: Dyslipidaemia with LDL-C derangement was highly prevalent, especially in individuals with both diabetes mellitus and hypertension, and there was absent use of lipid-lowering therapy. Screening and managing dyslipidaemia should be reinforced to reduce the risk of cardiovascular complications in this population at increased risk.

An audit of Heart failure management among ambulatory adult patients at Queen Elizabeth Central Hospital (QECH), Malawi.

Background: There are limited data on the clinical characteristics and use of guideline directed medical therapy among patients with heart failure in Malawi. We conducted a study to assess patient characteristics and clinical management given to heart failure patients at Queen Elizabeth Central hospital in Malawi. Methods: In a cross sectional study, patients with a diagnosis of heart failure who were followed up in the adult chest clinic at QECH were recruited to ascertain their characteristics and the therapy they were receiving. Echocardiograms and electrocardiograms were performed to identify abnormalities. Results: A total of 79 patients were recruited and 62% (49 out of 79) were female. The median age was 60 years (IQR 40.5-70.5). Most patients were hypertensive with NYHA (New York Heart Association) class I and II symptoms. Left ventricular(LV) systolic dysfunction was found in 55% (36 out of 65), with 68% (39 out of 65) having features of left ventricular remodeling. Most patients were on at least a single neurohormonal drug with 77% (61 out of 79) on ACEI (angiotensin converting enzyme inhibitor), 52% (42 out of 79) on a beta blocker and 34%(27 out of 79) on aldosterone antagonists. The recommended doses of medications were achieved in 14% (9 out 61), 24% (10 out 42), 22% (6 out of 27) on ACEI, beta blockers and aldosterone antagonists respectively. Conclusions: Hypertension is the commonest comorbidity in patients with heart failure, who are mostly females with NYHA class I or II symptoms. Most had LV remodeling changes and are on at least one neurohormonal antagonist but most remain sub optimally treated.

The prevalence of Schistosoma mansoni infection among adults with chronic non-communicable diseases in Malawi.

BACKGROUND: Schistosomiasis is a parasitic infectious disease caused by flatworms of the Schistosoma genus. The global burden of schistosomiasis is high. In Malawi, schistosomiasis is among the top 20 causes of outpatient department visits in health facilities. Schistosomiasis is among the most important but neglected causes of non-communicable diseases (NCD) peculiar to tropical endemic settings. While much is known about the contribution of S. haematobium to the NCD burden in Malawi, the role of S. mansoni remains largely unknown. METHODS: We conducted a cross-sectional study at Mangochi District Hospital. Adults over 18 years diagnosed with NCDs (n = 414), admitted or attending weekly outpatient clinics were recruited between August 2021 and February 2022. Data were collected on sociodemographic characteristics, medical history, body weight, blood pressure, and fasting blood glucose. Stool and midstream urine were collected for Kato-Katz (KK) microscopy and urine point of care-circulating cathodic antigen (POC-CCA) tests, respectively. We computed prevalence of S. mansoni as number of positive KK and CCA tests, each divided by total submitted samples. Univariate and multivariable logistic regression were done to evaluate risk factors of NCDs and association between S. mansoni infection and NCDs. RESULTS: We recruited 414 participants, mean age 57 years (SD 16), 67% of whom were female. Prevalence of S. mansoni based on urine CCA was 15% (95% CI: 11-19) and 0% on KK microscopy. Hypertension was the most common condition with a prevalence of 85% (95% CI: 81-89), followed by diabetes mellitus with a prevalence of 42% (95% CI: 37-46) and heart disease with a prevalence of 3% (95% CI: 2-5). S. mansoni infection was not significantly associated with hypertension (OR: 1.2, 95% CI: 0.5-3.1), diabetes (OR: 0.6, 95% CI: 0.3-1.10) or heart disease (OR: 2.0, 95% CI: 0.4-10). CONCLUSIONS: We observed moderate prevalence of S. mansoni infection among adults in the study per WHO classification of endemicity. This is within the range observed in children in Mangochi from 10 to 56.7%.

Meeting demand-Obstetric hemorrhage and blood availability in Malawi, a qualitative study.

BACKGROUND: Postpartum haemorrhage (PPH) is the leading cause of maternal mortality in Malawi. Despite the presence of a centralized institution supplying blood and blood products for hospitals across the country, a lack of timely blood transfusion has been identified as a critical barrier to successful PPH management. This study aims to understand the factors that affect the blood delivery pipeline and adequate access to blood products for postpartum haemorrhage patients. METHODS: Qualitative data were collected through in-depth interviews with key stakeholders across the blood delivery pipeline. Interviews were conducted from July 2020 to January 2021 at Queen Elizabeth Central Hospital and Mulanje District Hospital, a referral and district hospital respectively, as well as the Malawi Blood Transfusion Service. Line by line, open coding was used to perform a thematic analysis of the data using Nvivo and Atlas.ti software. RESULTS: Five key themes were identified: 1) Lack of blood availability due to an inadequate donor pool, 2) Transportation of blood products and PPH patients is impeded by distance to target sites and competing interests for blood delivery vehicles, 3) The Malawi Blood Transfusion Service has difficulty meeting demand for blood products due to inadequate funding and difficulty retaining blood donors, 4) Current PPH management protocols and practices lead to delays due to inconsistent guidelines on delivery and analysis of patient samples, and 5) Communication between health cadres is inconsistent and affected by a lack of adequate resources. CONCLUSIONS: Barriers to timely blood transfusion for PPH patients exist across the blood delivery pipeline. While an investment of infrastructure would alleviate many obstacles, several solutions identified in this study can be implemented without additional resources, such as establishing joint department meetings to improve communication between health cadres. Ultimately, given a resource limited setting, it may be worth considering de-centralizing the blood supply.

The WHO-TDR Dissemination and Implementation Massive Open Online Course (MOOC): Evaluation and Lessons Learned from Eight Low-and Middle-Income Countries.

Introduction: Non-communicable diseases (NCDs) are a leading cause of morbidity and mortality in low-and middle-income countries (LMICs). Despite this, a lack of funding, training and mentorship for NCD investigators in LMICs exists. In an effort to gain knowledge and skills to address these gaps, participants from the Global Research on Implementation and Translation Science (GRIT), a consortium of studies in eight LMICs and their networks, attended the dissemination and implementation (D&I) massive open online course (MOOC) developed by the Special Programme for Research and Training in Tropical Diseases at the World Health Organization to strengthen D&I capacity building. Here, we report on the feasibility of this MOOC, which was implemented during the SARS COVID-19 pandemic from April- November 2020. Methods: Participants completed pre- and post- training questionnaires to assess self-reported D&I competencies, general research skills, and research mentor access and quality. D&I competencies were measured by use of a scale developed for a US-based training program, with change in competency scores assessed by paired t test. We used univariate statistics to analyze the data for all other outcomes. Results: Of the 247 participants enrolled, 32 (13%) completed all MOOC components. D&I competency scores suggest improvement for those who had complete pre- and post-assessments. Trainee's average score on the full competency scale improved 1.45 points (0-5 scale) from pre- to post-test; all four subscales also showed evidence of improvements. There were small but not significant increases in competencies for grant writing, proposal/ manuscript writing and presentations from pre- to post-test assessment. 40% of trainees reported access to a research mentor and 12% reported access to a D&I specific mentor. Participants reported barriers (e.g., unstable internet access and challenges due to COVID-19) and facilitators (e.g., topical interests, collaboration with colleagues) to completing the MOOC. Conclusions: Although COVID-19 affected program usage and completion, the MOOC was feasible and effective, showing that among LMIC participants completing the course, there was improvement in D&I competency scores. Recommendations for future D&I trainings in LMICs should include 1) adding more topic specific modules (i.e., NCD research, general research skills) for scalability; 2) fostering more collaboration with participants across LMICs; and 3) establishing partnerships with D&I mentors for course participants.

Isoniazid-associated pellagra during mass scale-up of tuberculosis preventive therapy: a case-control study.

BACKGROUND: Pellagra is caused by niacin (vitamin B3) deficiency and patients with pellagra present with a characteristic rash. Isoniazid disrupts intracellular niacin synthesis and might induce niacin deficiency. In 2017, Malawi scaled up continuous isoniazid preventive treatment (IPT) for tuberculosis prevention among people living with HIV. In addition, an under-diversified diet based on subsistence maize, as is commonly the case in Malawi, is a risk factor for pellagra. We aimed to investigate whether large-scale isoniazid exposure in Malawi contributed to the cumulative risk for pellagra in a nutritionally vulnerable population. METHODS: We did a matched case-control study to evaluate the association between daily, continuous isoniazid exposure and pellagra. We matched sequentially enrolled patients with pellagra each with four control participants by sex and age from referral dermatology centres in three IPT scale-up districts in Malawi (Lilongwe, Blantyre, and Zomba) to evaluate isoniazid as a risk for pellagra using multivariable conditional logistic regression. We established a community clinic referral system surrounding the dermatology clinic in each district to enhance case-finding and included all patients with pellagra, regardless of referral status. The primary outcome was dermatologist-diagnosed pellagra. We calculated the interval between isoniazid initiation and rash onset and assessed 30-day clinical outcomes after multi-B vitamin treatment containing 300 mg nicotinamide daily. FINDINGS: Between Feb 5 and Aug 9, 2019, we enrolled 197 patients with pellagra and 781 matched controls. Isoniazid exposure was associated with an increased risk of pellagra (adjusted odds ratio 42·6 [95% CI 13·3-136·6]). Significant covariates included HIV infection, referral status, food insecurity, underweight, excess alcohol consumption, and, among women, lactation. The median time from isoniazid initiation to rash onset was shorter during the season of food scarcity (5 months [IQR 3-7]) compared with the harvest season (9 months [8-11]; hazard ratio 7·2 [95% CI 3·2-16·2], log-rank p<0·0001). Those with isoniazid-associated pellagra who discontinued isoniazid and adhered to multi-B vitamin treatment showed 30-day clinical improvement. INTERPRETATION: Continuous IPT scale-up and the annual period of food scarcity both increased the risk of pellagra in Malawi. Use of shorter rifamycin-based regimens for tuberculosis prevention and food fortification in populations with undernutrition might reduce this risk. Niacin-containing multi-B vitamin co-administration with isoniazid as pellagra prevention is worth exploring further. FUNDING: This study was supported by the President's Emergency Plan for AIDS Relief through the US Centers for Disease Control and Prevention under project 7173.

An insight into patients' perspectives on barriers affecting participation in shared decision making among patients with diabetes mellitus in Malawi.

BACKGROUND: Patient participation in decision making is a basic tenet for a patient centred care experience and, has potential to improve care experiences and responsiveness in chronic diseases such as Diabetes Mellitus (DM). However, documented experiences show that patient participation in decisions making is wanting. As Malawi strives to institutionalise patient centred care delivery, it is important to examine patients' experiences and perceptions to identify barriers affecting their participation in shared decision making because this may provide evidence supporting strategies in implementation of the institutionalisation. AIM: The study sought to describe perspectives about barriers to participation in shared decision making among patients with DM in Malawi. METHODS: This was an exploratory qualitative study. We targeted patients attending DM clinics in four public health facilities in southern Malawi from September to December 2019. We used In-Depth Interviews and Focus Group Discussions. Data was managed using Nvivo version 11 software and analysed using Content Analysis. RESULTS: The study highlights the values, perceptions and benefits of shared decision making. Furthermore, patients' narratives expose the struggles and vulnerabilities in their attempts to engage their providers towards shared decision making. CONCLUSION: Interactional power imbalances, insufficient dialogue and patients' own restrictive attitudes towards engagement with their providers thwarts SDM in clinical encounters. To make SDM a reality, transforming medical education that emphasizes on the value of good patient-provider relationship and providers' attitudes to regard patients as active partners may be a good starting point. Additionally, strategies that empower and change patients' perceptions about SDM require investment.

In-kitchen aerosol exposure in twelve cities across the globe.

Poor ventilation and polluting cooking fuels in low-income homes cause high exposure, yet relevant global studies are limited. We assessed exposure to in-kitchen particulate matter (PM2.5 and PM10) employing similar instrumentation in 60 low-income homes across 12 cities: Dhaka (Bangladesh); Chennai (India); Nanjing (China); Medellín (Colombia); São Paulo (Brazil); Cairo (Egypt); Sulaymaniyah (Iraq); Addis Ababa (Ethiopia); Akure (Nigeria); Blantyre (Malawi); Dar-es-Salaam (Tanzania) and Nairobi (Kenya). Exposure profiles of kitchen occupants showed that fuel, kitchen volume, cooking type and ventilation were the most prominent factors affecting in-kitchen exposure. Different cuisines resulted in varying cooking durations and disproportional exposures. Occupants in Dhaka, Nanjing, Dar-es-Salaam and Nairobi spent > 40% of their cooking time frying (the highest particle emitting cooking activity) compared with âˆ¼ 68% of time spent boiling/stewing in Cairo, Sulaymaniyah and Akure. The highest average PM2.5 (PM10) concentrations were in Dhaka 185 ± 48 (220 ± 58) µg m-3 owing to small kitchen volume, extensive frying and prolonged cooking compared with the lowest in Medellín 10 ± 3 (14 ± 2) µg m-3. Dual ventilation (mechanical and natural) in Chennai, Cairo and Sulaymaniyah reduced average in-kitchen PM2.5 and PM10 by 2.3- and 1.8-times compared with natural ventilation (open doors) in Addis Ababa, Dar-es-Salam and Nairobi. Using charcoal during cooking (Addis Ababa, Blantyre and Nairobi) increased PM2.5 levels by 1.3- and 3.1-times compared with using natural gas (Nanjing, Medellin and Cairo) and LPG (Chennai, Sao Paulo and Sulaymaniyah), respectively. Smaller-volume kitchens (<15 m3; Dhaka and Nanjing) increased cooking exposure compared with their larger-volume counterparts (Medellin, Cairo and Sulaymaniyah). Potential exposure doses were highest for Asian, followed by African, Middle-eastern and South American homes. We recommend increased cooking exhaust extraction, cleaner fuels, awareness on improved cooking practices and minimising passive occupancy in kitchens to mitigate harmful cooking emissions.