Population health status surveillance and its determinants in Bagira Health Zone in the Democratic Republic of the Congo: Analysis and prospects

Technical Summary: Population Health Status Surveillance and its Determinants in Bagira Health Zone, DRC

Problem Statement
Public health surveillance is a cornerstone of essential public health activities, yet in the Democratic Republic of the Congo (DRC), existing systems often fail to capture the full spectrum of health determinants, particularly in urban-rural zones like Bagira. While previous studies in South Kivu have examined healthcare access or specific disease burdens (e.g., malaria in children), there is a lack of comprehensive analysis regarding the general determinants of population health status and how these factors interact within an urban-rural context. Furthermore, informal health facilities and the specific sociodemographic and behavioral nuances of slum and peri-urban populations are frequently absent from national data systems and policy priorities. The study addresses the gap in understanding the specific drivers of health status in the Bagira Health Zone (HZ) and evaluates whether current surveillance mechanisms adequately account for these determinants.

Methodology
The research employed a cross-sectional survey design conducted between July and September 2023 in the Bagira HZ, Bukavu.

• Population and Sampling: The study targeted 505 adults (≥18 years) attending health facilities or residing in the community. A sample size of 424 was calculated based on a 95% confidence level and 50% prevalence assumption, expanded to 505 to account for non-responses. Sampling utilized the law of large numbers to ensure normal distribution.
• Data Collection: Data were gathered using three distinct questionnaires administered via KoboCollect:
  1. For community members (sick and non-sick) to assess sociodemographics, economic status, and health behaviors.
  2. For patients and non-patients in integrated health centers and the Bagira Referral General Hospital to determine care-seeking behaviors.
  3. For HZ management staff to evaluate existing surveillance levels.
• Health Status Measurement: The primary dependent variable was the population health status, measured using the EQ-5D Health-Related Quality of Life (HRQoL) tool. This included the EQ Visual Analogue Scale (EQ-VAS) for self-rated health (0–100) and the calculation of an "Unhealthy Days Index" (sum of poor physical and mental health days, capped at 30).
• Statistical Analysis: Data were processed using Stata 14. Descriptive statistics summarized variables. Bivariate analysis (simple linear regression) identified associations with a threshold of p < 0.25. Significant variables were entered into a multiple linear regression model to identify independent predictors of health status. Collinearity was checked using Variance Inflation Factors (VIF < 3), and missing data were handled using the XGBoost algorithm.

Key Results

• Demographics and Economics: The sample (N=505) had a mean age of 33.5 years, with 57.4% female and 66.9% belonging to the Shi tribe. The majority were self-employed (62.1%), with a mean monthly household income of USD 79.9. Most lived in semi-permanent housing (65.0%).
• Healthcare Access and Utilization: Health centers were the most utilized (80.6%) and preferred (69.7%) facilities. Preventive care (vaccination, education) was the most frequently offered service (95.6%). However, 31.5% of respondents had been treated for a medical problem in the HZ, and waiting times averaged 73.1 minutes.
• Health Status Perception: Only 43.4% of respondents rated their general health as "good." The mean EQ-VAS score was 64.1% (SD 17.6), indicating an "average" perceived health status. The median Unhealthy Days Index was 11.0 days (IQR 6.0–21.0). Respondents reported an average of 7.0 days of poor physical health and 7.0 days of poor mental health in the past 30 days.
• Determinants of Health Status: Multiple linear regression identified several significant factors influencing health status:
  • Sociodemographic: Being male (β = 0.126, p = 0.026) and belonging to the Shi tribe (β = -0.003, p < 0.001) were significant determinants.
  • Healthcare Provision: Attendance at private health facilities (β = 0.115, p = 0.004), the affordability of care (β = -0.096, p = 0.019), and preferring the General Referral Hospital (β = 0.113, p = 0.001) were associated with health status.
  • Individual Behavior: Being a smoker (β = -0.139, p = 0.010) and having been treated for a medical problem (β = 0.001, p = 0.001) were significant.
  • Mental and Physical Dimensions: The study highlighted that mental and physical health dimensions are critical components of HRQoL, with pain and depression significantly limiting daily activities for a large portion of the population.

Significance and Claims
The paper claims that the health of the Bagira HZ population is multifactorially influenced by sociodemographic characteristics, healthcare provision factors, individual behaviors, and crucially, the mental and physical dimensions of quality of life.

A primary conclusion is that the existing surveillance system in the Bagira HZ does not currently take these specific determinants into account. The authors argue that current monitoring mechanisms are insufficient for capturing the complex interplay of factors affecting urban-rural health in this context.

The study's significance lies in its attempt to derive a theoretical basis for monitoring health status in an urban-rural HZ by identifying specific predictive factors. It suggests that to improve surveillance, systems must be interconnected across health zones and integrated with broader disease monitoring to include these identified determinants. The authors emphasize that mental health programs and economic development are necessary to address the identified gaps in HRQoL, noting that a multidisciplinary approach is required to manage the cluster of comorbidities affecting the population.