A Multilevel Regression Analysis Protocol for Evaluating Urban Primary Care Network Performance and Clinical Outcomes in Kenya
Fuente:
Zenodo
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Recurso digital |
| Sprache: | Englisch |
| Veröffentlicht: |
Zenodo
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866901362723258368 |
|---|---|
| author | Hassan, Fatuma Okoth, Omondi Mwangi, Wanjiku |
| author_facet | Hassan, Fatuma Okoth, Omondi Mwangi, Wanjiku |
| contents | <p>{ "background": "Urban primary care networks (PCNs) are a critical component of health system strengthening in sub-Saharan Africa, yet robust methodologies for evaluating their performance and impact on clinical outcomes are lacking. Existing evaluations often fail to account for the hierarchical structure of data inherent in networked care delivery.", "purpose and objectives": "This protocol details a methodological approach for evaluating the performance of urban PCNs in Kenya. The primary objective is to quantify the association between PCN-level structural and process factors and individual-level clinical outcomes for hypertension and type 2 diabetes, while controlling for patient and facility characteristics.", "methodology": "We propose a multilevel regression analysis using a two-level hierarchical model. Individual patient outcomes (level-1) are nested within PCNs (level-2). The core statistical model is: $y{ij} = \\beta{0j} + \\beta{1}X{ij} + \\epsilon{ij}$, where $\\beta{0j} = \\gamma{00} + \\gamma{01}Z{j} + u{0j}$. Here, $y{ij}$ is the clinical outcome for patient $i$ in PCN $j$, $X{ij}$ are patient-level covariates, $Z{j}$ are PCN-level predictors, and $u{0j}$ is the PCN-specific random effect. Inference will be based on 95% confidence intervals and robust standard errors. Data will be extracted from routine health information systems and a linked facility assessment survey.", "findings": "As a research protocol, this paper does not present empirical results. The anticipated findings will include estimated coefficients quantifying the direction and magnitude of PCN predictors on clinical outcomes. For example, we hypothesise that a higher proportion of facilities within a network achieving a minimum service readiness score will be associated with a clinically significant improvement in patient-level disease control rates.", "conclusion": "The proposed methodology provides a rigorous, generalisable framework for evaluating networked primary care systems. Its application will yield evidence on the</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18951880 |
| institution | Zenodo |
| language | eng |
| publishDate | 2022 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | A Multilevel Regression Analysis Protocol for Evaluating Urban Primary Care Network Performance and Clinical Outcomes in Kenya Hassan, Fatuma Okoth, Omondi Mwangi, Wanjiku Primary care networks Multilevel regression analysis Sub-Saharan Africa Health systems evaluation Clinical outcomes Kenya Urban health systems <p>{ "background": "Urban primary care networks (PCNs) are a critical component of health system strengthening in sub-Saharan Africa, yet robust methodologies for evaluating their performance and impact on clinical outcomes are lacking. Existing evaluations often fail to account for the hierarchical structure of data inherent in networked care delivery.", "purpose and objectives": "This protocol details a methodological approach for evaluating the performance of urban PCNs in Kenya. The primary objective is to quantify the association between PCN-level structural and process factors and individual-level clinical outcomes for hypertension and type 2 diabetes, while controlling for patient and facility characteristics.", "methodology": "We propose a multilevel regression analysis using a two-level hierarchical model. Individual patient outcomes (level-1) are nested within PCNs (level-2). The core statistical model is: $y{ij} = \\beta{0j} + \\beta{1}X{ij} + \\epsilon{ij}$, where $\\beta{0j} = \\gamma{00} + \\gamma{01}Z{j} + u{0j}$. Here, $y{ij}$ is the clinical outcome for patient $i$ in PCN $j$, $X{ij}$ are patient-level covariates, $Z{j}$ are PCN-level predictors, and $u{0j}$ is the PCN-specific random effect. Inference will be based on 95% confidence intervals and robust standard errors. Data will be extracted from routine health information systems and a linked facility assessment survey.", "findings": "As a research protocol, this paper does not present empirical results. The anticipated findings will include estimated coefficients quantifying the direction and magnitude of PCN predictors on clinical outcomes. For example, we hypothesise that a higher proportion of facilities within a network achieving a minimum service readiness score will be associated with a clinically significant improvement in patient-level disease control rates.", "conclusion": "The proposed methodology provides a rigorous, generalisable framework for evaluating networked primary care systems. Its application will yield evidence on the</p> |
| title | A Multilevel Regression Analysis Protocol for Evaluating Urban Primary Care Network Performance and Clinical Outcomes in Kenya |
| topic | Primary care networks Multilevel regression analysis Sub-Saharan Africa Health systems evaluation Clinical outcomes Kenya Urban health systems |
| url | https://doi.org/10.5281/zenodo.18951880 |