Browsing by Subject "Sub-Saharan Africa"

Background: Nutrition is a key determinant of pediatric cancer patient outcomes in sub-Saharan Africa. Accurately identifying pediatric cancer patients at risk of malnutrition remains a challenge. There is a need for a standardized nutritional screening tool, developed and validated in sub-Saharan Africa. Study aims: The aims of this study were to: 1) select candidate variables in the development of a nutritional screening from predictors associated with malnutrition in pediatric cancer patients and, 2) conduct a secondary data analysis estimating the prevalence of pediatric cancer in Uganda from cases presented at the Uganda Cancer Institute between January 1, 2017 and December 31, 2019. Methods: This study is a longitudinal hospital-based study, carried out at the Bugando Medical Center in Tanzania and Uganda Cancer Institute in Uganda. The study enrolled clinically confirmed pediatric cancer patients (<18 years) at the study sites. Measures of interest include: nutritional status, symptom duration, abdominal distention, anthropometric measures such as height, weight, mid-upper arm circumference, abdominal circumference, triceps skinfold thickness, and clinical characteristics such as serum albumin, mean corpuscular volume, and protein. Logistic regression models examined predictors of nutritional status in pediatric cancer patients. Lastly, geospatial analysis estimated the prevalence and examined the country-wide distribution of the pediatric cancers presented at the Uganda Cancer Institute between 2017 and 2019. Results: The sample of 77 pediatric cancer patients enrolled at the two study sites ranged from 1 to 17 years old. Solid tumor malignancies like Wilms tumor comprise of 40% of all diagnoses. 60% of cancer patients were malnourished at baseline. The strongest predictors of nutritional status were mid-upper arm circumference (AOR 0.52, 95% CI: 0.31 – 0.87), abdominal circumference (AOR 1.38, 95% CI: 1.16 – 1.65) and serum albumin (AOR 0.73, 95% CI: 0.62 - 0.86). Secondary analysis of the Uganda Cancer Institute registry shows 11607 patients with confirmed cancer diagnosis between 2017 and 2019. Acute lymphoblastic leukemia (31.4%) is the most common cancer diagnosis, followed by Wilms tumor (19.1%), rhabdomyosarcoma (9.4%) and Burkitt’s lymphoma (6.9%). Blood cancers are most common cancer types, of them the most frequent cases being leukemia (37%). 2018 saw the highest number of cancer presentations within the study timeframe. Conclusions: The results show abdominal circumference, serum albumin, and muac are candidate variables in developing a nutritional screening tool for pediatric cancer patients in SSA. Blood and solid cancers are prevalent in Uganda; thus, a customized nutritional screening tool is much needed.

This dissertation examines aspects of the relationship between improved energy access,both in terms of cooking energy and electricity access, and women’s time use patterns, labor productivity, and empowerment in low- and middle-income countries. The first chapter of this dissertation examines the relationship between women’s empowerment and various measures of cooking energy and electricity access across 7 countries in Sub-Saharan Africa and South and Southeast Asia using the multi-tiered framework datasets from the World Bank. Since there are many potential facets to women’s empowerment, for example, social standing (i.e., ability to participate in community groups, ability to move freely), employment, or education levels, we use principal component analysis to create an “empowerment index” that captures multiple aspects of women’s empowerment as a singular value. We then use simple regression analysis to study the correlation between women’s empowerment and energy access measures. We find positive associations between empowerment and measures of energy access, though this pattern is not consistent across all countries and contexts.

After descriptively establishing a positive relationship between women’s empowermentand improved cooking energy access, especially in Sub-Saharan African contexts, the second chapter of this dissertation describes an impact evaluation of an improved cookstove distributed in Tanzania, Malawi, and Zambia. We used a quasi-experimental design to survey approximately 3,000 households across three countries, looking for impacts on women’s time use patterns and labor productivity as a result of take up of the improved cookstove. Using a difference-indifferences approach, we find that in most contexts, this improved cookstove intervention does not result in changes to time use patterns, labor productivity, or time use agency, though the lack of positive impacts may be due to sample contamination, too short of a time frame between stove installation and endline surveys, or reporting errors in modules where time use data is collected.

In order to understand these results in the context of prior published evidence of timesavings from improved cookstoves, in the third chapter, we investigate the population and study characteristics that may impact the time saved in fuel collection as a result of the distributed improved cookstove. Specifically, we apply Bayesian linear regression modeling and Bayesian model comparison to investigate whether and how methodological and contextual choices, such as geography, level of remoteness of a region, fuel use behaviors, the type of time use elicitation method used, and respondent characteristics affect estimates of time savings in fuel collection derived from the cookstove distributed in Chapter 2. Our prior is constructed from 34 estimates of time savings from the improved cookstove literature, while our sampling data is provided by the quasi-experiment in Chapter 2. The approach provides insight on how different sources of variation impact time savings estimates and allows us to make predictions of potential time savings in new settings. Results suggest that the potential for time savings from this improved cookstove is highest in poorer, less educated populations.

In this dissertation, I contribute to the literature by first describing the relationshipsbetween forms of energy access, including improved cooking technologies, and women’s empowerment, and describing those patterns across countries. I then test this relationship using quasi-experimental methods to find causal impacts of improved cooking technologies on outcomes pertinent to women’s livelihoods, including women’s time use patterns, across four countries in Sub-Saharan Africa. Finally, I provide insight into how population and study characteristics impact time savings results from improved cooking technologies, and in what contexts we might find maximum impact.

BACKGROUND: This study describes the prehospital transport of traumatic brain injury (TBI) patients and its impact on TBI outcome to inform quality improvement for the existing trauma system. Data was collected over 4 months at a major referral hospital in Moshi,Tanzania.

METHODS: Patient demographics, mechanism of injury, injury severity (Glasgow Coma Score), and vitals were recorded on presentation to the Casualty Department. Prehospital factors recorded include time, distance and cost. Multivariable regression analyses evaluated the effect of prehospital factors on unfavourable patient TBI outcome, in-hospital factors and demographics were controlled for. Unfavorable outcome was defined as Glasgow Outcome Score<5 on discharge or death.

RESULTS: Road traffic injuries were the most common mechanism of injury (67.1%). The majority of patients were referred from other facilities in and around the region (62.3%), with 23% from the local public hospital There was no evidence of prehospital care available in this region. Average prehospital duration was more than 1 hour, a third of this was spent in prehospital transit for a majority of the patients. A minority used Ambulances. Predictors of unfavourable outcome (GOS<5) were: prehospital time greater than 60 minutes, multiple physical transfers during the prehospital course and being referred from another hospital.

CONCLUSION: The lack of prehospital care calls for further research into prehospital interventions for this setting. Further analysis should be conducted with a larger sample size to increase accuracy of the findings.

Background: Hypertension, a leading risk for cardiovascular mortality, is an important co-morbidity among people living with HIV (PLWH). Understanding the epidemiologic patterns of hypertension and barriers to hypertension care is a critical first step in developing interventions to improve cardiovascular outcomes among PLWH in sub-Saharan Africa (SSA). Methods: Semi-structured in-depth interviews were conducted with 13 hypertensive patients engaged in HIV care in northern Tanzania. Patients attending a single HIV clinic in Moshi, Tanzania were screened consecutively for hypertension. Hypertension was defined as having at least one of the following: 1) a self-reported diagnosis of hypertension, 2) a single blood pressure measurement > 160 mmHg systolic or > 100 mmHg diastolic, or 3) two measurements at separate visits, obtained at least one week apart of > 140 mmHg systolic or > 90 mmHg diastolic. All screened participants completed a short demographic survey. Participants who met criteria for hypertension were invited to complete a knowledge, attitudes and practices survey that included questions about previous hypertension history. Results: The data revealed a range of themes including poor understanding of hypertension causes, limited hypertension knowledge, perceived overlap and comparisons between hypertension and HIV, delays in hypertension diagnosis and linkage to care, shortcomings of provider communication and counseling, reluctance to use antihypertensive medication, lack of integration of hypertension and HIV care, and multiple barriers to hypertension care. Over an 8-week period, 555 patients were screened for hypertension. Hypertension prevalence was 19.6% (20.3% among women and 16.8% among men). The median age of patients with hypertension was 50 (IQR 43 – 57 years), compared to 42 (IQR 35 – 49 years) years for normotensive patients. Older age (OR 1.07, 95% CI 1.04 – 1.09, p<0.001) and higher body mass index (BMI) (OR 1.13, 95% CI 1.09 – 1.18, p<0.001) were independently associated with hypertension. Among the 91 patients with hypertension who completed additional questions about their hypertension history, 44 (48%) reported having a prior blood pressure check, 32 (35%) were previously aware of their hypertension diagnosis, and only 21 (23%) reported ever taking antihypertensive drugs. Of the 10 patients (11%) reporting current antihypertensive use, none had a controlled blood pressure. Conclusions: Participants described multiple, intersecting challenges related to hypertension management. Hypertension prevalence among patients engaged in HIV care was high. Despite a high burden of disease, the rates of screening, awareness, treatment and control were unacceptably low. Multifaceted strategies that address barriers to care along the hypertension care continuum and that are integrated within HIV care are urgently needed to improve cardiovascular outcomes among PLWH in SSA.

Background: The burden of cardiovascular disease (CVD) is rising in sub-Saharan Africa (SSA). Preeclampsia, a type of hypertensive disorder of pregnancy, is a unique risk factor for CVD among women, yet little is known about the postpartum cardiovascular risk among women with preeclampsia in SSA. Objective: To determine the incidence of hypertension, a major risk factor for CVD, at 6-months postpartum among Kenyan women with preeclampsia. Methods: This prospective cohort study included all pregnant or recently postpartum women with preeclampsia who were admitted to a national, referral hospital in western Kenya from January 20, 2020-March 19, 2020, when the study was unexpectedly paused due to the COVID-19 pandemic. Using home blood pressure monitoring technique, we described the trajectory of blood pressure after delivery. Bivariate and multivariable regression analyses were performed to investigate for risk factors associated with hypertension at 6-month follow-up. Results: Eight-six women with preeclampsia were enrolled prior to March 19, 2020, when the study was unexpected paused due to the COVID-19 pandemic. Among the 50 women who completed follow up, 38% (n=19) had hypertension. Blood pressure normalized for all women by 6 weeks after delivery but rose again beyond 12 weeks among those with hypertension at follow up. Maternal age, parity and history of preeclampsia in prior pregnancy or a previous pregnancy complication (preterm delivery or stillbirth) were associated with hypertension at follow up. Overall, rates of routine, postpartum clinic follow up were low (64%) among women despite nearly all (97%) attending clinic for infant immunizations following delivery. Conclusion: There is a high incidence of hypertension at 6-months postpartum among Kenyan women with preeclampsia, though low rates of postpartum follow up care indicate a potential missed opportunity for early CVD identification and prevention among this high risk-population.

Background: People living with HIV (PLWH) are at increased risk for ischemic heart disease (IHD) and myocardial infarction (MI). Our objectives in this study were the following: identify the prevalence of prior MI and myocardial ischemia among PLWH, compare with HIV-uninfected participants, identify correlates of prior MI among PLWH; examine PLWH knowledge, attitudes, and practices (KAP) regarding IHD, identify predictors of improved knowledge, and explore patient and provider perspectives on barriers to MI care. Methods: Adults presenting for routine HIV-care were recruited. Trained research assistants collected basic sociodemographic information and medical history and administered a standardized knowledge, attitudes, and practices survey regarding IHD. An IHD knowledge score between 0 and 10 was calculated for each participant based on the number of correct answers on the KAP knowledge questions. Associations between participant characteristics and IHD knowledge scores were assessed via Welch’s t-test. Age- and sex- matched participants without HIV were selected as controls from an outpatient department in the same district. Both PLWH and control participants underwent 12-lead resting ECG testing. ECG criteria for prior MI was defined as pathologic Q waves in ≥2 contiguous lead and myocardial ischemia as ST segment depression or T wave inversion in ≥2 contiguous leads. Each ECG was reviewed by two independent physician adjudicators, and Pearson’s chi-squared test was used to compare the prevalence of ischemic ECG findings among those with and without HIV. Multivariate logistic regression was also performed to identify patient characteristics correlating with prior MI in PLWH on ECG. Finally, PLWH with ECG findings of prior MI and physicians caring for PLWH were recruited via purposive sampling for in-depth interviews regarding barriers to MI care in PLWH. Inductive thematic analysis was conducted to identify emergent themes and all transcripts were independently coded in NVivo 12. Results: Of 500 enrolled PLWH, the mean (sd) age was 45.3 (11.4) years and 139 were (27.8%) males. The study found an average knowledge score of 5.6, with higher IHD knowledge scores being associated with post-primary education (mean = 6.27 vs 5.35, p = 0.001) and with >10% five-year risk of cardiovascular event (mean = 5.97 vs 5.41, p = 0.045). Control participants without HIV were recruited for 497 participants with HIV. Prior MI (11.1% vs 2.4%, OR 4.97, 95% CI: 2.71-9.89, p<0.001), and myocardial ischemia (18.7% vs 12.1% OR 1.67, 95% CI: 1.18-2.39, p=0.004) were significantly more common among participants with HIV than those without HIV. On multivariate analysis, prior MI among participants with HIV was associated with male sex (OR 2.05, 95% CI: 1.06-3.89, p=0.030) and self-reported family history of cardiovascular disease (OR 2.05, 95% CI: 1.05-3.88, p=0.031). The 24 in-depth interviews of providers and patients revealed barriers related to perceptions and understanding of MI among patients and providers, delays in MI diagnosis and initiation of treatment, and challenges in the provision of long-term MI care. Conclusions: Tanzanians with HIV have a significantly higher burden of IHD than those without HIV, but significant barriers exist to high quality cardiovascular care in this population. Further study is required to characterize cardiovascular disease burden and improve care for PLWH.

Background: Acute coronary syndrome (ACS) is a rare diagnosis in sub-Saharan Africa, but patient and physician-related factors that may contribute to under-diagnosis are poorly understood. Methods: A community survey was performed of adults in northern Tanzania, and respondents were asked to describe symptoms of ACS, personal healthcare-seeking preferences for chest pain, and perceptions of self-risk. A retrospective chart review was conducted at a referral hospital in northern Tanzania to determine rates of various diagnoses among admitted adults over a six-month period. A prospective observational study was conducted in the emergency department, and diagnostic workups, treatments, and diagnoses for adults with chest pain or shortness of breath were recorded. Results: Of 718 survey respondents, 277 (38.6%) stated they would present to a hospital for chest pain and 115 (16.0%) were able to identify a conventional ACS symptom. Retrospectively, ACS accounted for 9 (0.3%) adult admissions, whereas heart failure accounted for 294 (12.2%) admissions. Prospectively, among 339 adults presenting with chest pain or shortness of breath, 170 (50.1%) received an electrocardiogram, 9 (2.7%) underwent cardiac biomarker testing, and 6 (1.8%) were diagnosed with ACS. Conclusions: In northern Tanzania, community awareness of ACS symptoms is low and only a minority of adults would present to a hospital for chest pain. Full diagnostic workups are rare among patients with symptoms of possible ACS, and ACS is a rare diagnosis.

Background Globally, it is estimated that 11% of all disability adjusted life years lost result from conditions requiring surgical intervention. Efforts to estimate burden at country-specific levels have been recommended to plan surgical delivery platforms. However, existing analyses of hospital records are not representative of population-level needs. Therefore, we piloted a population-level epidemiologic survey in a large, peri-urban District of Uganda. The exercise would inform implementation of the eventual nation-wide survey of Uganda.

Methods A 2-stage cluster sampling design was used to sample fifty five (55) households. In each household, up to 2 individual respondents were recruited. Village Health Team (VHT) members served as enumerators and used the Surgeons OverSeas Assessment of Surgical Need (SOSAS) instrument to acquire self-reported data on existing surgical conditions and surgical history. A head/representative of household was asked about household deaths within the previous 12-months. Descriptive statistics, weighted adjustments, and regression modeling were used to analyze results.

Results Six of 96 individual respondents (6.25%) reported an existing surgical condition. The lifetime prevalence of surgical conditions was 26.0% (25 of 96 individuals), reporting 33 total surgical conditions. The most commonly reported problems were wound related (54.5% - 18 of 33). The most common anatomic regions affected were face/head/neck, extremities, and abdomen. Injuries were responsible for 51.5% of reported surgical conditions. Two of three household deaths involved proximate causes that were surgically treatable. For all met and unmet need, 80% (20 of 25 individuals) were treated or need treatment at a District Hospital or lower level facility.

Prevalence of existing surgical conditions was used in this pilot to compute a nation-wide study sample size of 4,750. The pilot study cost USD 25/respondent and averaged 36 minutes per household. Major revisions in deployment of the nation-wide survey included: hiring enumerators who lived in the EAs, random household sampling, shift in data collection equipment, and improving breath and depth of data acquired by the SOSAS instrument.

Conclusion The prevalence of existing surgical conditions in Wakiso District was within range of previous pilot studies deploying the same SOSAS instrument, albeit not from Uganda. A large proportion of individuals have surgical problems that can be treated at District Hospital or lower level facilities. The pilot study was feasible and necessary to inform deployment of the nation-wide survey.

Background: Sepsis is a leading cause of death and disability globally. Despite a high burden of sepsis in sub-Saharan Africa, clinical data for sepsis in that setting are limited. We sought to describe the clinical characteristics, management, and outcomes in a cohort of adults and adolescents with sepsis in northern Tanzania. We also assessed for associations between clinical factors and in-hospital mortality.

Methods: We carried out a prospective observational cohort study at Kilimanjaro Christian Medical Centre in Moshi, Tanzania. We collected data on demographics, baseline clinical characteristics, and management, with an emphasis on hours 0-6 after arrival to the Emergency Department. Log risk regression was carried out to assess for associations between demographic and clinical factors and our primary outcome of in-hospital death. Separate multivariable regression analyses were conducted for both antimicrobial administration by hour 6 and administration of intravenous (IV) fluids >1L by hour 6 and the outcome of in-hospital mortality.

Results: Fifty-eight participants were included in our analysis. Seventeen (29.3%) participants died in-hospital. Baseline characteristics associated with inpatient mortality included inability to drink unassisted, respiratory rate >30 breaths per minute, hypoxia, and altered mentation. Less than half of participants received any antimicrobial by hour 6, and most participants received <1L of IV fluids. HIV antibody testing was performed for only one participant in the first 6 hours. On multivariable analysis, neither antimicrobial administration nor IV fluids >1L by hour 6 was associated with inpatient mortality.

Conclusion: Sepsis in northern Tanzania carries a high risk of in-hospital mortality. Further research is urgently needed to establish the highest-yield interventions suited to the unique characteristics of sepsis in sSA.

Background: Evidence mainly from high-income countries has demonstrated sex related differences in the incidence, presentation and management of patients with non-valvular AF. Such evidence is scarce in sub-Saharan Africa, yet there is a rising prevalence of AF. This study aimed to determine sex differences in the distribution and treatment pattern for modifiable risk factors in Western Kenya.

Methods: The study included two phases. Phase 1 comprised of secondary data analysis from a case control study – Study of Genetics of Atrial Fibrillation in an African population. Phase 2 included retrospective analysis of medical records at the cardiac clinic in a large referral hospital in Kenya. We determined the distribution and treatment pattern for modifiable risk factors for AF using chi-square and fisher’s exact test.

Results: Hypertension is the most prevalent modifiable risk factor for AF in western Kenya. The prevalence among men and women was 65% and 76% respectively, but this difference was not significant. Three percent of men were obese compared to 24 % of women (p =0.013). Men were more likely to drink alcohol (p = 0.001) and have a history of smoking compared to women (p = <0.001). Among men, tetra choric correlation showed a very strong association between smoking and alcohol intake (correlation coefficient > 0.9), and hypertension and obesity/overweight (correlation coefficient >0.9). These correlations were weaker among women with a correlation coefficient of 0.40 and 0.38 respectively. Among the participants, only 21% had weight and height measurements recorded. Nutritional counselling was recorded for only 3% of those who had a BMI > 29.9. Similarly, less than 10% of those with a history of smoking or alcohol intake received counselling on cessation strategies.

Conclusion: Hypertension is the most common modifiable risk factor for AF in western Kenya. There are significant differences among men and women in the distribution of dyslipidemia, alcohol intake, smoking and obesity. These modifiable risk factors have strikingly low rates of interventions. Management of patients with AF should include both screening and interventions for modifiable risk factors. Packaging of intervention should consider sex-specific differences.

Introduction

Methods

Results

Discussion

Abstract

Background: Fractional exhaled nitric oxide (FeNO) is a novel biomarker that is utilized as a tool to assist with the diagnosis and management of asthma in developed countries. Little data exists from sub-Saharan Africa to understand the role for exhaled nitric oxide in subjects suspected of having asthma. In this study, we aim to elucidate if a relationship exists between elevated FeNO levels and symptoms of asthma.

Methods: Using a cluster randomized stratified sampling strategy, 154 subjects age 12 and above in Uasin Gishu County, Kenya were enrolled. Questionnaires including ISAAC written and video questionnaires and the St George’s Respiratory Questionnaire for COPD were completed. Subjects were tested for FeNO, pre-and post-bronchodilator spirometry, and exhaled carbon monoxide. Odds ratios for the presence of asthma symptoms based on FeNO levels above or below a cutoff point of 71 ppb were generated.

Results: Overall, 5.8% of subjects were identified with asthma symptoms by video questionnaires, and 17.5% by written questionnaires. Median FeNO levels were significantly higher in subjects with wheezing compared to those without. The odds ratio for wheezing in individuals with FeNO levels greater than 71 ppb compared to less than 71 ppb was 7.8 (video questionnaire) and 11.3 (written questionnaire).

Conclusions: A statistically significant relationship exists between elevated levels of FeNO and symptoms of asthma in this western Kenyan population. Further work is needed to explore this link for clinical and research purposes.

Background: Traditional medicine use is increasingly recognized as a common and important component of healthcare globally. Our study aim was therefore to identify the commonly used traditional medicines in Moshi, Tanzania, the factors influencing their use and associations between traditional medicine use & prevalence of chronic diseases. Methods: We performed a secondary data analysis of a mixed methods study in Moshi, comprising 42 extended interviews and 5 focus group discussions with key informants, and cross-sectional household survey using interviewer-administered questionnaires and field-based diagnostic tests for CKD, diabetes, hypertension and HIV. Results: We identified 168 traditional medicines, of which 15 (8.9%) and 5 (3%) were used to treat chronic diseases and CKD, respectively. Participants reported seeking healthcare advice from medical doctors (97%), family members (52%), pharmacists (24%) and friends or neighbors (14%). In a fully adjusted model, CKD patients were more likely than the non-CKD population to report a history of traditional medicine use (AOR=1.99; p=0.04), and family tradition (OR=1.97), difficulty finding a medical doctor (OR=2.07) and fewer side effects with traditional medicines (OR=2.07) as their reasons for preferring traditional medicines to hospital medicines. Conclusions: Traditional medicine use is high in Moshi, and more so among the CKD population. A history of traditional medicine use is associated with the prevalence of CKD in Moshi. Most of these traditional medicines have biologically active substances that could potentially be developed into therapeutic and prophylactic therapies for CKD, and CKD-associated co-morbidities.

Background: Routine immunization is the most effective way to prevent mortality from vaccine preventable diseases. Though vaccination coverage rates have improved over the past decade, vaccinations still fail to be delivered and received within a proper timeframe. Across low and middle income countries the median vaccination delay falls between 2.3 to 6.2 weeks for birth through third dose vaccinations [1]. Countries in the African region have 5 to 6 percent lower coverage and face lower timeliness than LMICs overall, and in Tanzania, vaccination delays reach up to 70 percent [1, 2]. This study seeks to investigate the determinants of timely vaccination delivery using two approaches. First, a review is conducted to assess the causes of delayed vaccination in Sub-Saharan Africa (SSA) and, second, an analysis of the Tanzania Demographic and Health Surveys (DHS) data provides evidence of the timeliness factors of concern. Methods: Peer-reviewed literature on vaccination coverage and timeliness in SSA was searched on several databases and was subsequently selected if it pertained to the determinants of vaccination coverage or vaccination timeliness for humans. Pharmacological interventions and vaccine efficacy research was excluded. The literature is rigorously reviewed by classification of determinants into four framework clusters (family characteristics, immunization systems, parental knowledge/ attitudes, and communication and information). The 2015 Tanzania DHS data is used to measure timely immunization. Determinants of fully timely immunization are analyzed through logistic regression. Results: The results of the literature review provided 36 papers on vaccination timeliness and 15 papers on vaccination timeliness. Family characteristics are found to be the major determinants of both vaccination coverage and timeliness appearing 29 and 6 times, respectively. The immunizations system was also found to impact vaccination coverage and vaccination timeliness with 24 and 3 appearances. Our DHS analysis supports these findings, showing statistically significant links between complete timeliness and family characteristics (maternal age, wealth quintile, maternal education, and region) and the immunization system (antenatal care visits and tetanus toxoid vaccine). Conclusions: This investigation finds several family characteristics and the immunization system as determinants of delayed vaccination delivery. Escalating the number of vaccination facilities or health facilities in underserved regions could alleviate disparities found amongst families, as these characteristics are often clustered regionally. Further, action must be taken to improve the immunization system by ensuring availability of vaccines everywhere, good vaccine management, and adequate staff. Understanding the causes of poor vaccination timeliness is crucial in efforts to improve problems and provide timely vaccination to prevent child mortality.

Background: There is a rising burden of myocardial infarction (MI) within sub-Saharan Africa. Prospective studies of detailed MI outcomes in the region are lacking.

Methods:Adult patients with confirmed MI from a prospective surveillance study in northern Tanzania were enrolled in a longitudinal cohort study after baseline health history, medication use, barriers to care, and sociodemographics were obtained. Thirty days following hospital presentation, symptom status, rehospitalizations, medication use, and mortality were assessed via telephone or in-person interviews using a standardized follow-up questionnaire. Multivariate logistic regression was performed to identify baseline predictors of thirty-day survival.

Results:Thirty-day follow-up was achieved for 150 (98.7%) of 152 enrolled participants. Of these, 85 (56·7%) survived to thirty-day follow-up. Of the surviving participants, 71 (83·5%) reported persistent anginal symptoms, four (4·7%) reported taking aspirin regularly, seven (8·2%) were able to identify MI as the reason for their hospitalization, and 17 (20·0%) had unscheduled rehospitalizations. Baseline predictors of thirty-day survival included self-reported history of diabetes (OR 0·32, 95% CI 0·10-0·89, p = 0·04), self-reported history of hypertension (OR 0·34, 95% CI 0·15-0·74, p = 0·01) and antiplatelet use at initial presentation (OR 0·19, 95% CI 0·04-0·65, p = 0·02).

Conclusions:In northern Tanzania, thirty-day outcomes following acute MI are poor, and mortality is associated with comorbidities and medication usage. Further investigation is needed to develop interventions to improve care and outcomes of MI in Tanzania.