Pairwise fitting of piecewise mixed models for the joint modeling of multivariate longitudinal outcomes, in a randomized crossover trial.

Many statistical models have been proposed in the literature for the analysis of longitudinal data. One may propose to model two or more correlated longitudinal processes simultaneously, with a goal of understanding their association over time. Joint modeling is then required to carefully study the association structure among the outcomes as well as drawing joint inferences about the different outcomes. In this study, we sought to model the associations among six nutrition outcomes while circumventing the computational challenge posed by their clustered and high-dimensional nature. We analyzed data from a 2 × $\times$ 2 randomized crossover trial conducted in Kenya, to compare the effect of high-dose and low-dose iodine in household salt on systolic blood pressure (SBP) and diastolic blood pressure (DBP) in women of reproductive age and their household matching pair of school-aged children. Two additional outcomes, namely, urinary iodine concentration (UIC) in women and children were measured repeatedly to monitor the amount of iodine excreted through urine. We extended the model proposed by Mwangi et al. (2021, Communications in Statistics: Case Studies, Data Analysis and Applications, 7(3), 413-431) allowing flexible piecewise joint models for six outcomes to depend on separate random effects, which are themselves correlated. This entailed fitting 15 bivariate general linear mixed models and deriving inference for the joint model using pseudo-likelihood theory. We analyzed the outcomes separately and jointly using piecewise linear mixed-effects (PLME) model and further validated the results using current state-of-the-art Jones and Kenward methodology (JKME model) used for analyzing randomized crossover trials. The results indicate that high-dose iodine in salt significantly reduced blood pressure (BP) compared to low-dose iodine in salt. Estimates for the random effects and residual error components showed that SBP and DBP had strong positive correlation, with effect of the random slope indicating that significantly related outcomes are strongly associated in their evolution. There was a moderately strong inverse relationship between evolutions of UIC and BP both in women and children. These findings confirmed the original hypothesis that high-dose iodine salt has significant lowering effect on BP. We further sought to evaluate the performance of our proposed PLME model against the widely used JKME model, within the multivariate joint modeling framework through a simulation study mimicking a 2 × 2 $2\times 2$ crossover design. From our findings, the multivariate joint PLME model performed exceptionally well both in estimation of random-effects matrix (G) and Hessian matrix (H), allowing satisfactory model convergence during estimation. It allowed a more complex fit to the data with both random intercepts and slopes effects compared to the multivariate joint JKME model that allowed for random intercepts only. When a hierarchical viewpoint is adopted, in the sense that outcomes are specified conditionally upon random effects, the variance-covariance matrix of the random effects must be positive definite. In some cases, additional random effects could explain much variability in the data, thus improving precision in estimation of the estimands (effect size) parameters. The key highlight in this evaluation shows that multivariate joint JKME model is a powerful tool especially while fitting mixed models with random intercepts only, in crossover design settings. Addition of random slopes may lead to model complexities in most cases, resulting in unsatisfactory model convergence during estimation. To circumvent convergence pitfalls, extention of JKME model to PLME model allows a more flexible fit to the data (generated from crossover design settings), especially in the multivariate joint modeling framework.

Evaluation of a flexible piecewise linear mixed-effects model in the analysis of randomized cross-over trials.

Cross-over designs are commonly used in randomized clinical trials to estimate efficacy of a new treatment. They have received a lot of attention, particularly in connection with regulatory requirements for new drugs. The main advantage of using cross-over designs over conventional parallel designs is increased precision, thanks to within-subject comparisons. In the statistical literature, more recent developments are discussed in the analysis of cross-over trials, in particular regarding repeated measures. A piecewise linear model within the framework of mixed effects has been proposed in the analysis of cross-over trials. In this article, we report on a simulation study comparing performance of a piecewise linear mixed-effects (PLME) model against two commonly cited models-Grizzle's mixed-effects (GME) and Jones & Kenward's mixed-effects (JKME) models-used in the analysis of cross-over trials. Our simulation study tried to mirror real-life situation by deriving true underlying parameters from empirical data. The findings from real-life data confirmed the original hypothesis that high-dose iodine salt have significantly lowering effect on diastolic blood pressure (DBP). We further sought to evaluate the performance of PLME model against GME and JKME models, within univariate modeling framework through a simulation study mimicking a 2 × 2 cross-over design. The fixed-effects, random-effects and residual error parameters used in the simulation process were estimated from DBP data, using a PLME model. The initial results with full specification of random intercept and slope(s), showed that the univariate PLME model performed better than the GME and JKME models in estimation of variance-covariance matrix (G) governing the random effects, allowing satisfactory model convergence during estimation. When a hierarchical view-point is adopted, in the sense that outcomes are specified conditionally upon random effects, the variance-covariance matrix of the random effects must be positive-definite. The PLME model is preferred especially in modeling an increased number of random effects, compared to the GME and JKME models that work equally well with random intercepts only. In some cases, additional random effects could explain much variability in the data, thus improving precision in estimation of the estimands (effect size) parameters.

Analysis of user interactions with a digital health wallet for enabling care continuity in the context of an ongoing pandemic.

BACKGROUND: The onset of COVID-19 and related policy responses made it difficult to study interactive health informatics solutions in clinical study settings. Instrumented log and event data from interactive systems capture temporal details that can be used to generate insights about care continuity during ongoing pandemics. OBJECTIVE: To investigate user interactions with a digital health wallet (DHW) system for addressing care continuity challenges in chronic disease management in the context of an ongoing pandemic. MATERIALS AND METHODS: We analyzed user interaction log data generated by clinicians, nurses, and patients from the deployment of a DHW in a feasibility study conducted during the COVID-19 pandemic in Kenya. We used the Hamming distance from Information Theory to quantify deviations of usage patterns extracted from the events data from predetermined workflow sequences supported by the platform. RESULTS: Nurses interacted with all the user interface elements relevant to triage. Clinicians interacted with only 43% of elements relevant to consultation, while patients interacted with 67% of the relevant user interface elements. Nurses and clinicians deviated from the predetermined workflow sequences by 42% and 36%, respectively. Most deviations pertained to users going back to previous steps in their usage workflow. CONCLUSIONS: User interaction log analysis is a valuable alternative method for generating and quantifying user experiences in the context of ongoing pandemics. However, researchers should mitigate the potential disruptions of the actual use of the studied technologies as well as use multiple approaches to investigate user experiences of health technology during pandemics.

Isolation, identification and associated risk factors of non-tuberculous mycobacteria infection in humans and dromedary camels in Samburu County, Kenya.

Non-tuberculous mycobacteria are of public health significance, and zoonotic infection is attributed to the sociocultural practice of consumption of raw milk and the close human-livestock contact in pastoral communities. This study aimed at isolation, identification of mycobacteria from human sputum and camel milk and risk factors assessment in Samburu East, Kenya. Six hundred and twelve camels and 48 people presumed to have tuberculosis (TB) from 86 households in Wamba and Waso regions were screened. Camels were categorized into Somali, Turkana and Rendile breeds. Single intradermal comparative tuberculin test (SICTT) was used as a herd-screening test on lactating camels and a milk sample collected from reactive camels. Sputum samples were collected from eligible members of participating households. A standard questionnaire on possible risk factors for both humans and camels was administered to respective household heads or their representatives. Total camel skin test reactors were 238/612 (38.9%). Milk and sputum samples were analysed at KEMRI/TB research laboratory for microscopy, GeneXpert® , culture and identification. Isolates were identified using 16S rRNA gene sequencing at Inqaba biotec in South Africa. Sixty-four isolates were acid-fast bacilli (AFB) positive of which M. fortuitum (3), M. szulgai (20), M. monacense (5), M. lehmanni (4), M. litorale (4), M. elephantis (3), M. duvalii (3), M. brasiliensis (1), M. arcueilense (1) and M. lentiflavum (1) were from milk; M. fortuitum (1), M. szulgai (2) and M. litorale (1) were from humans. Risk factors included the following: Turkana breed (OR = 3.4; 95% CI: 1.2-9.3), replacements from outside the County (OR = 2.1; 95% CI: 0.3-12.3), presence of other domestic species (small stock; OR = 4.6) and replacement from within the herd (OR = 3.2; 95% CI: 0.7-14.7). Zoonotic risk practices included raw milk consumption, shared housing and handling camels. Monitoring of zoonotic NTM through surveillance and notification systems is required.

Factors associated with occurrence of salmonellosis among children living in Mukuru slum, an urban informal settlement in Kenya.

BACKGROUND: In Kenya, typhoid fever and invasive non-typhoidal salmonellosis present a huge burden of disease, especially in poor-resource settings where clean water supply and sanitation conditions are inadequate. The epidemiology of both diseases is poorly understood in terms of severity and risk factors. The aim of the study was to determine the disease burden and spatial distribution of salmonellosis, as well as socioeconomic and environmental risk factors for these infections, in a large informal settlement near the city of Nairobi, from 2013 to 2017. METHODS: Initially, a house-to-house baseline census of 150,000 population in Mukuru informal settlement was carried out and relevant socioeconomic, demographic, and healthcare utilization information was collected using structured questionnaires. Salmonella bacteria were cultured from the blood and faeces of children < 16 years of age who reported at three outpatient facilities with fever alone or fever and diarrhea. Tests of association between specific Salmonella serotypes and risk factors were conducted using Pearson Chi-Square (χ2) test. RESULTS: A total of 16,236 children were recruited into the study. The prevalence of bloodstream infections by Non-Typhoidal Salmonella (NTS), consisting of Salmonella Typhimurium/ Enteriditis, was 1.3%; Salmonella Typhi was 1.4%, and this was highest among children < 16 years of age. Occurrence of Salmonella Typhimurium/ Enteriditis was not significantly associated with rearing any domestic animals. Rearing chicken was significantly associated with high prevalence of S. Typhi (2.1%; p = 0.011). The proportion of children infected with Salmonella Typhimurium/ Enteriditis was significantly higher in households that used water pots as water storage containers compared to using water directly from the tap (0.6%). Use of pit latrines and open defecation were significant risk factors for S. Typhi infection (1.6%; p = 0.048). The proportion of Salmonella Typhimurium/ Enteriditis among children eating street food 4 or more times per week was higher compared to 1 to 2 times/week on average (1.1%; p = 0.032). CONCLUSION: Typhoidal and NTS are important causes of illness in children in Mukuru informal settlement, especially among children less than 16 years of age. Improving Water, Sanitation and Hygiene (WASH) including boiling water, breastfeeding, hand washing practices, and avoiding animal contact in domestic settings could contribute to reducing the risk of transmission of Salmonella disease from contaminated environments.

A cross-sectional epidemiological investigation of nontuberculous mycobacteria of public health importance in slaughter camels in Samburu County, Kenya.

The diagnosis of tuberculosis (TB) in camels at slaughter houses heavily relies on post mortem (PM) meat inspection to detect granulomatous lesions; however, the sensitivity of this technique is not perfect. The objective of this study was to isolate and characterize mycobacteria associated with suspect TB pathological lesions at PM. At PM, 1600 camels were examined in two county slaughterhouses. One hundred and thirty two, 8.25% (132/1600) (Binomial CI 95% 6.95-9.71%), suspect granulomatous lesions were found. Twenty seven, 1.69% (27/1600) (Binomial CI 95% 1.11-2.45%), were confirmed as acid-fast bacilli (AFB) using Ziehl-Neelsen (ZN) staining after culture. Speciation using the GenoType® Mycobacterium assay (Hain Lifesciences, Nehren, Germany) found a majority isolates to be Mycobacterium fortuitum (17), the other species identified included M. szulgai (2), M. scrofulaceum (3), M. marinum (1), M. intracellulare (1), M. gordonae (1), and 2 unidentified mycobacteria species. The types of lesions observed were nodular, caseous masses involving whole organs or cavities, and purulent masses. The highest proportion of suspect lesions were observed in the right, left bronchial lymph nodes, and the mediastinal lymph nodes (59.54%), followed by the retropharyngeal lymph nodes (12.21%), the medial lobe (10.67%), and the left lateral and quadrate lobes of the lungs (17.58%). The 6-7 age category had higher odds (OR = 2.5) of culture positivity. It was concluded that a variety of NTM species of medical importance were associated with TB lesions in the thoracic lymph nodes and lungs. There is need to unravel the public health significance of these mycobacteria.

Intrinsic and extrinsic factors associated with sputum characteristics of presumed tuberculosis patients.

BACKGROUND: Sputum remains the most preferred specimen for detection of Mycobacterium tuberculosis due to its non-invasive method of production. Good quality sputum specimen is essential for accurate diagnosis of pulmonary tuberculosis (PTB). It is therefore imperative to assess factors that are related to the production of sputum that is of the best quality. OBJECTIVE: We assessed the intrinsic and extrinsic characteristics of presumed tuberculosis patients and the quality of sputum they produced. METHODS: This was a cross-sectional study in which consenting enrolled presumed tuberculosis patients were subjected to medical examination and a structured questionnaire administered to collect clinical history, demographic information, environmental and behavioral characteristics. The enrolled participants were instructed on how to collect spot and morning sputum specimens for macroscopic and microscopic assessment to determine any association. RESULTS: A total of 309 patients were enrolled into the study with an even distribution on gender (50.5% males). Of these, 202 (65.3%) submitted both a spot and a morning specimen for analysis. On macroscopic examination, 70% spot and 68% morning sputum were characterized as good quality (Purulent/mucoid). The factors associated (p<0.05) with quality specimen included both intrinsic and extrinsic factors. The intrinsic factors included: difficulty in breathing, presence of conjunctivitis and knowledge of the disease whereas the only extrinsic factor associated with production of good quality sputum for tuberculosis diagnosis was time taken by patient to seek tuberculosis treatment after occurrence of any of the TB symptoms. CONCLUSION: Both intrinsic and extrinsic factors affected the quality of sputum produced by presumed tuberculosis patients. Clinical and behavioral characteristics including conjunctivitis, difficulty in breathing and delay in seeking treatment were important factors that determined the production of good quality sputum specimens, while knowledge of tuberculosis disease did not compel presumed tuberculosis patients to produce good quality sputum for diagnosis of the disease.

A study to determine the prevalence and factors associated with hypertension among employees working at a call centre Nairobi Kenya.

INTRODUCTION: Hypertension often referred to as Non Communicable Diseases (NCDs). Causes of hypertension are classified into modifiable and non-modifiable factors. The objective of the study was to determine the prevalence and other associated factors leading to the onset of hypertension among employees working at the call center. METHODS: This was a descriptive cross sectional study design. Data collection was done in two parts; part one comprised of clinical health assessments; weight and height to aid determine Body Mass Index and blood pressure measurement. Part two was by self-administered questionnaires to participants to aid identify behavioral risk factors and further elicit lifestyle practices. Data was collected from a sample population of 370 respondents. Descriptive statistical analysis was applied in univariate analysis. Further analysis included bivariate and multiple regression analysis; Odds Ratio with 95% confidence interval was used to determine the strength of association. RESULTS: The proportion of hypertension was significantly higher among overweight respondents (32.7%) (OR= 11.55; 95% CI= 4.44-30.07; P < 0.001) and obese respondents (60.2%) (OR= 36.02; 95% CI= 13.43-96.60; P < 0.001) compared to those respondents who were within normal range of weight (4.0%). Nine (9) factors that were associated with hypertension at bivariate analysis (P < 0.05) were all subjected to a multiple regression analysis or reduced model where four factors remained in the final analysis. Respondents who were classified as overweight had 10.6 times likelihood developing hypertension compared to those respondents with normal weight (AOR= 10.61; 95%CI= 3.98-28.32; P < 0.001). Likewise, obese respondents were 43.6 fold more likely to develop hypertension compared to those respondents within normal range of weight [OR=43.68; 95%CI=15.24-125.16; P<0.001]. Respondents not trying to reduce fat in their diet were highly predisposed having hypertension at (AOR=2.44; 95% CI=1.20-4.96; P= 0.014) than respondents who always tried to reduce fat in their diet. Respondents who sometimes engage on more physical exercises were 2.2 times likely to develop hypertension (AOR=2.22; 95%CI= 1.20-4.10; P= 0.011) compared to those who always engaged in more physical exercises. Respondents with parenting issues were about twice as likely to have hypertension (AOR= 2.15; 95% CI: 1.23-3.74; P= 0.007) than parents who did not have parenting issues. CONCLUSION: This study depicts rising cases of hypertension and an alarming rate of pre-hypertension among the working population. This vary based on the age, obesity, parental responsibility, unhealthy diet and lack of or reduced physical activity. These call for strategic interventions and greater emphasis on health promotion programs at the workplace alongside staff empowerment towards health seeking behaviors.

Nutritional status of cancer outpatients using scored patient generated subjective global assessment in two cancer treatment centers, Nairobi, Kenya.

BACKGROUND: Malnutrition is a universal problem in cancer patients renowned as an important factor for increased morbidity, decreased quality of life and high mortality. Early diagnosis of malnutrition risk through nutrition screening followed by comprehensive and timely interventions reduces mortality associated with malnutrition. The Scored Patient-Generated Subjective Global Assessment (PGSGA) method has been proved efficient in identifying cancer patients with nutrition challenges and guiding appropriate interventions. However this tool has not been adopted in management of cancer patients in Kenya. The aim of the study was to assess and describe nutrition status of cancer outpatients receiving treatment at Kenyatta National Hospital Hospital (KNH) and Texas Cancer Centre (TCC). METHODS: The study adopted a hospital based descriptive cross sectional study. Cancer outpatients with confirmed stage 1-4 cancers, physically stable, aged 18 years and above and receiving cancer treatment were recruited and assessed using Scored PGSGA tool. Proportions, measures of central tendency and pearsons' chi-square test were used in statistical analysis. RESULTS: Among the 471 participants assessed, 71.8% were female and 28.2% male. Most participants had stage 2, 3 and 4 cancers at 27.2%, 27.2% and 24.3% respectively. Highest proportion of participants had breast (29.7%) and female genital cancers (22.9%). Sixty nine percent of participants were well nourished (SGA-A), 19.7% moderately malnourished (SGA-B) and 11.3% severely malnourished (SGA-C) and this difference was statistically significant. The mean PGSGA score was 6.76 (SD 5.17). Based on the score, 33.8% of participants required critical nutrition care, 34.8% symptoms management, 14.2% constant nutrition education and pharmacological intervention while 17.2% required routine assessments and reassurance. More (m;54.7%, f; 45.3%) males than females were severely malnourished(SGA-C) and this was statistically significant (P < 0.001).Prevalence of severe malnutrition was highest among participants with digestive organ cancers (49.1%) followed by those with lip cancer (17%) and the least prevalence reported in those with Karposi Sarcoma (0%). Most of stage 4 participants were moderately (37.5%) and severely (29.4%) malnourished. CONCLUSIONS: The Scored Patient-Generated Subjective Global Assessment is able to identify cancer patients both at risk of malnutrition and those severely malnourished. It also provides a guideline on the appropriate nutrition intervention hence an important tool in nutrition management of cancer patients.

Factors associated with adequate weekly reporting for disease surveillance data among health facilities in Nairobi County, Kenya, 2013.

INTRODUCTION: Kenya adopted the Integrated Disease Surveillance and Response (IDSR) strategy in 1998 to strengthen disease surveillance and epidemic response. However, the goal of weekly surveillance reporting among health facilities has not been achieved. We conducted a cross-sectional study to determine the prevalence of adequate reporting and factors associated with IDSR reporting among health facilities in one Kenyan County. METHODS: Health facilities (public and private) were enrolled using stratified random sampling from 348 facilities prioritized for routine surveillance reporting. Adequately-reporting facilities were defined as those which submitted >10 weekly reports during a twelve-week period and a poor reporting facilities were those which submitted <10 weekly reports. Multivariate logistic regression with backward selection was used to identify risk factors associated with adequate reporting. RESULTS: From September 2 through November 30, 2013, we enrolled 175 health facilities; 130(74%) were private and 45(26%) were public. Of the 175 health facilities, 77 (44%) facilities classified as adequate reporting and 98 (56%) were reporting poorly. Multivariate analysis identified three factors to be independently associated with weekly adequate reporting: having weekly reporting forms at visit (AOR19, 95% CI: 6-65], having posters showing IDSR functions (AOR8, 95% CI: 2-12) and having a designated surveillance focal person (AOR7, 95% CI: 2-20). CONCLUSION: The majority of health facilities in Nairobi County were reporting poorly to IDSR and we recommend that the Ministry of Health provide all health facilities in Nairobi County with weekly reporting tools and offer specific trainings on IDSR which will help designate a focal surveillance person.

Using the unmet obstetric needs indicator to map inequities in life-saving obstetric interventions at the local health care system in Kenya.

BACKGROUND: Developing countries with high maternal mortality need to invest in indicators that not only provide information about how many women are dying, but also where, and what can be done to prevent these deaths. The unmet Obstetric Needs (UONs) concept provides this information. This concept was applied at district level in Kenya to assess how many women had UONs and where the women with unmet needs were located. METHODS: A facility based retrospective study was conducted in 2010 in Malindi District, Kenya. Data on pregnant women who underwent a major obstetric intervention (MOI) or died in facilities that provide comprehensive Emergency Obstetric Care (EmOC) services in 2008 and 2009 were collected. The difference between the number of women who experienced life threatening obstetric complications and those who received care was quantified. The main outcome measures in the study were the magnitude of UONs and their geographical distribution. RESULTS: 566 women in 2008 and 724 in 2009 underwent MOI. Of these, 185 (32.7%) in 2008 and 204 (28.1%) in 2009 were for Absolute Maternal Indications (AMI). The most common MOI was caesarean section (90%), commonly indicated by Cephalopelvic Disproportion (CPD)-narrow pelvis (27.6% in 2008; 26.1% in 2009). Based on a reference rate of 1.4%, the overall MOI for AMI rate was 1.25% in 2008 and 1.3% in 2009. In absolute terms, 22 (11%) women in 2008 and 12 (6%) in 2009, who required a life saving intervention failed to get it. Deficits in terms of unmet needs were identified in rural areas only while urban areas had rates higher than the reference rate (0.8% vs. 2.2% in 2008; 0.8% vs. 2.1% in 2009). CONCLUSIONS: The findings, if used as a proxy to maternal mortality, suggest that rural women face higher risks of dying during pregnancy and childbirth. This indicates the need to improve priority setting towards ensuring equity in access to life saving interventions for pregnant women in underserved areas.

Field-derived Schistosoma mansoni and Biomphalaria pfeifferi in Kenya: a compatible association characterized by lack of strong local adaptation, and presence of some snails able to persistently produce cercariae for over a year.

BACKGROUND: Schistosoma mansoni is widely distributed in sub-Saharan Africa with Biomphalaria pfeifferi being its most widespread and important snail intermediate host. Few studies have examined the compatibility of field-derived B. pfeifferi snails with S. mansoni miracidia derived from human hosts. We investigated compatibility (as defined by shedding of cercariae following exposure to miracidia) of two isolates of S. mansoni from school children from Asao (western Kenya) and Mwea (central Kenya) with B. pfeifferi collected directly from Asao stream or the Mwea rice fields. METHODS: We exposed snails from both regions to four different doses of miracidia (1, 5, 10 and 25) from sympatric or allopatric S. mansoni, and maintained them in a shaded, screened out-of-doors rearing facility in Kisian, in western Kenya. Both snail survival and the number of snails that became infected were monitored weekly. This was done for 25 weeks post-exposure (PE). Those infected snails which survived beyond this period were monitored until they all died. RESULTS: Although overall survival of Mwea snails maintained in western Kenya was generally low, both sympatric and allopatric combinations of parasites and snails exhibited high compatibility (approximately 50% at a dose of one miracidium per snail), with an increase in infection rates as the miracidial dose was increased (P < 0.002). Schistosomes were no more compatible with sympatric than allopatric snails, nor were snails less compatible with sympatric than allopatric schistosomes. Snail mortality increased significantly with dose of miracidia (P < 0.05). Approximately 3% of Asao snails exposed to a low dose of sympatric miracidia (1 or 5) continued to shed cercariae for as long as 58 weeks post exposure. CONCLUSIONS: There were no significant local adaptation effects for either schistosomes or snails. Also, the existence of "super-survivor" snails is noteworthy for its implications for current control initiatives that mostly rely on mass drug administration (MDA). Long-term shedders could provide an ongoing source of cercariae to initiate human infections for many months, suggesting care is required in considering how human MDA treatments are timed. Future control programs should incorporate means to eliminate infected snails to complement chemotherapy interventions in controlling schistosomiasis.

Food Insecurity and Not Dietary Diversity Is a Predictor of Nutrition Status in Children within Semiarid Agro-Ecological Zones in Eastern Kenya.

Machakos and Makueni counties in Kenya are associated with historical land degradation, climate change, and food insecurity. Both counties lie in lower midland (LM) lower humidity to semiarid (LM4), and semiarid (LM5) agroecological zones (AEZ). We assessed food security, dietary diversity, and nutritional status of children and women. Materials and Methods. A total of 277 woman-child pairs aged 15-46 years and 6-36 months respectively, were recruited from farmer households. Food security and dietary diversity were assessed using standard tools. Weight and height, or length in children, were used for computation of nutritional status. Findings. No significant difference (P > 0.05) was observed in food security and dietary diversity score (DDS) between LM4 and LM5. Stunting, wasting, and underweight levels among children in LM4 and LM5 were comparable as were BMI scores among women. However, significant associations (P = 0.023) were found between severe food insecurity and nutritional status of children but not of their caregivers. Stunting was significantly higher in older children (>2 years) and among children whose caregivers were older. Conclusion. Differences in AEZ may not affect dietary diversity and nutritional status of farmer households. Consequently use of DDS may lead to underestimation of food insecurity in semiarid settings.

Existence and functionality of emergency obstetric care services at district level in Kenya: theoretical coverage versus reality.

BACKGROUND: The knowledge on emergency obstetric care (EmOC) is limited in Kenya, where only partial data from sub-national studies exist. The EmOC process indicators have also not been integrated into routine health management information system to monitor progress in safe motherhood interventions both at national and lower levels of the health system. In a country with a high maternal mortality burden, the implication is that decision makers are unaware of the extent of need for life-saving care and, therefore, where to intervene. The objective of the study was to assess the actual existence and functionality of EmOC services at district level. METHODS: This was a facility-based cross-sectional study. Data were collected from 40 health facilities offering delivery services in Malindi District, Kenya. Data presented are part of the "Response to accountable priority setting for trust in health systems" (REACT) study, in which EmOC was one of the service areas selected to assess fairness and legitimacy of priority setting in health care. The main outcome measures in this study were the number of facilities providing EmOC, their geographical distribution, and caesarean section rates in relation to World Health Organization (WHO) recommendations. RESULTS: Among the 40 facilities assessed, 29 were government owned, seven were private and four were voluntary organisations. The ratio of EmOC facilities to population size was met (6.2/500,000), compared to the recommended 5/500,000. However, using the strict WHO definition, none of the facilities met the EmOC requirements, since assisted delivery, by vacuum or forceps was not provided in any facility. Rural-urban inequities in geographical distribution of facilities were observed. The facilities were not providing sufficient life-saving care as measured by caesarean section rates, which were below recommended levels (3.7% in 2008 and 4.5% in 2009). The rates were lower in the rural than in urban areas (2.1% vs. 6.8%; p < 0.001 ) in 2008 and (2.7% vs. 7.7%; p < 0.001) in 2009. CONCLUSIONS: The gaps in existence and functionality of EmOC services revealed in this study may point to the health system conditions contributing to lack of improvements in maternal survival in Kenya. As such, the findings bear considerable implications for policy and local priority setting.

Prevalence of metabolic syndrome among an urban population in Kenya.

OBJECTIVE: Developing countries are undergoing an epidemiologic transition accompanied by increasing burden of cardiovascular disease (CVD) linked to urbanization and lifestyle modifications. Metabolic syndrome is a cluster of CVD risk factors whose extent in Kenya remains unknown. The aim of this study was to determine the prevalence of metabolic syndrome and factors associated with its occurrence among an urban population in Kenya. RESEARCH DESIGN AND METHODS: This was a household cross-sectional survey comprising 539 adults (aged ≥18 years) living in Nairobi, drawn from 30 clusters across five socioeconomic classes. Measurements included waist circumference, HDL cholesterol, triacylglycerides (TAGs), fasting glucose, and blood pressure. RESULTS: The prevalence of metabolic syndrome was 34.6% and was higher in women than in men (40.2 vs. 29%; P < 0.001). The most frequently observed features were raised blood pressure, a higher waist circumference, and low HDL cholesterol (men: 96.2, 80.8, and 80%; women: 89.8, 97.2, and 96.3%, respectively), whereas raised fasting glucose and TAGs were observed less frequently (men: 26.9 and 63.3%; women: 26.9 and 30.6%, respectively). The main factors associated with the presence of metabolic syndrome were increasing age, socioeconomic status, and education. CONCLUSIONS: Metabolic syndrome is prevalent in this urban population, especially among women, but the incidence of individual factors suggests that poor glycemic control is not the major contributor. Longitudinal studies are required to establish true causes of metabolic syndrome in Kenya. The Kenyan government needs to create awareness, develop prevention strategies, and strengthen the health care system to accommodate screening and management of CVDs.

Delivery practices and associated factors among mothers seeking child welfare services in selected health facilities in Nyandarua South District, Kenya.

BACKGROUND: A measure of the proportion of deliveries assisted by skilled attendants is one of the indicators of progress towards achieving Millennium Development Goal (MDG) 5, which aims at improving maternal health. This study aimed at establishing delivery practices and associated factors among mothers seeking child welfare services at selected health facilities in Nyandarua South district, Kenya to determine whether mothers were receiving appropriate delivery care. METHODS: A hospital-based cross-sectional survey among women who had recently delivered while in the study area was carried out between August and October 2009. Binary Logistic regression was used to identify factors that predicted mothers' delivery practice. RESULTS: Among the 409 mothers who participated in the study, 1170 deliveries were reported. Of all the deliveries reported, 51.8% were attended by unskilled birth attendants. Among the deliveries attended by unskilled birth attendants, 38.6% (452/1170) were by neighbors and/or relatives. Traditional Birth Attendants attended 1.5% (17/1170) of the deliveries while in 11.7% (137/1170) of the deliveries were self administered. Mothers who had unskilled birth attendance were more likely to have <3 years of education (Adjusted Odds ratio [AOR] 19.2, 95% confidence interval [CI] 1.7 - 212.8) and with more than three deliveries in a life time (AOR 3.8, 95% CI 2.3 - 6.4). Mothers with perceived similarity in delivery attendance among skilled and unskilled delivery attendants were associated with unsafe delivery practice (AOR 1.9, 95% CI 1.1 - 3.4). Mother's with lower knowledge score on safe delivery (%) were more likely to have unskilled delivery attendance (AOR 36.5, 95% CI 4.3 - 309.3). CONCLUSION: Among the mothers interviewed, utilization of skilled delivery attendance services was still low with a high number of deliveries being attended by unqualified lay persons. There is need to implement cost effective and sustainable measures to improve the quality of maternal health services with an aim of promoting safe delivery and hence reducing maternal mortality.