Application of Principal Component Analysis to Assess Health Systems Capacity Using Cross Sectional Data in Rural Western Kenya

Maximila N. Wanzala^1, , J. A Oloo¹, Gordon Nguka¹ and Vincent Were²

¹Department of Public Health, Masinde Muliro University of Science and Technology

²Kenya Medical Research Institute, Centre for Global Health Research, Kisumu, Kenya

Cite this paper

Maximila N. Wanzala, J. A Oloo, Gordon Nguka and Vincent Were. Application of Principal Component Analysis to Assess Health Systems Capacity Using Cross Sectional Data in Rural Western Kenya. American Journal of Public Health Research. 2019; 7(1):27-32. doi: 10.12691/AJPHR-7-1-5

Abstract

Introduction: Strong health systems are essential platforms for accessible, quality health services, and population health and attainment of the Sustainable Development Goal (SDGs). Descriptive methods have been used to assess the health systems strength and impact, however, there is inadequate knowledge on methods of analyzing huge number of indices to provide systematic evidence that service readiness is improving or deteriorating over time. Methods: We utilized data from a cross section survey of 71 health facilities in Kakamega County of western Kenya. A total of 151 indices of the health system building blocks were reduced using Principal Component Analysis (PCA) model which generated factor weights for the individual indicators. These included indices from human resources, service delivery, infrastructure, finance, health information systems, commodities and governance. Factors weights were then summed and ranked in order of their relative contribution to better performance. These were then summed and average to rank health facilities. Sum of indicators within each health system block was used as explanatory variables in a linear regression model with overall average of all indicators. Coefficients of the regression was used to assess marginal effects and p-value<0.05 were considered statistics significant. Results: The top ranked indicators were basic service deliver for testing and diagnosis and the lowest ranked were infrastructure such as availability of public taps, water, toilet or privacy. The department that were highly ranked whose indicators performed better in terms of weighting, were service delivery (p<0.0001), health financing (p<0.0001), health workforce (p=0.005) and medical supplies and commodities (p<0.0001) in relation to overall service provision denoted by overall weighting for all indicators. Health governance was not a significant factor influencing service provision. Conclusion: PCA is an essential methodology for assessing health system readiness and preparedness to provide accessible and quality service delivery in resource poor settings.

Keywords

principal component analysis, health systems, devolution

Copyright

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

References

[1]	De Savigny D, Adam T. Systems thinking for health systems strengthening. World Health Organization; 2009.
[2]	Williamson T, Mulaki A. Devolution of Kenya’s health system, the role of HPP. RTI International. 2015.
[3]	Wambua KC, Kiruthu Z. Decentralization of Government Operations and Service Delivery Performance by County Governments in Kenya. Master's Thesis, University of Nairobi, School of Business, Nairobi, Kenya. 2014.
[4]	Government of Kenya G. Kenya Service Availability and Readiness Assessment Mapping (SARAM). Nairobi Ministry of Health; 2014.
[5]	Tsofa B, Molyneux S, Gilson L, Goodman C. How does decentralisation affect health sector planning and financial management? a case study of early effects of devolution in Kilifi County, Kenya. International journal for equity in health. 2017; 16(1): 151.
[6]	Govenment of Kenya G. Health Survey 2014: key indicators. Kenya National Bureau of Statistics (KNBS) and ICF Macro. 2014.
[7]	Kolenikov S, Angeles G. The use of discrete data in PCA: theory, simulations, and applications to socioeconomic indices. Chapel Hill: Carolina Population Center, University of North Carolina. 2004:1-59.
[8]	Kolenikov S, Angeles G. The use of discrete data in principal component analysis for socio-economic status evaluation. University of North Carolina at Chapel Hill, Carolina, NC. [Online] February. 2005.
[9]	Vyas S, Kumaranayake L. Constructing socio-economic status indices: how to use principal components analysis. Health policy and planning. 2006; 21(6): 459-468.
[10]	World_Health_Organization. (WHO). Handbook on health inequality monitoring with a special focus on low-and middle-income countries. WHO; 2013.
[11]	Jackson EF, Siddiqui A, Gutierrez H, Kanté AM, Austin J, Phillips JF. Estimation of indices of health service readiness with a principal component analysis of the Tanzania Service Provision Assessment Survey. BMC health services research. 2015; 15(1): 536.
[12]	Health Mo. Kenya Health Policy 2014-2030. Ministry of Health Nairobi, Kenya; 2014.
[13]	Kenya_Ministry_of_Health_(MOH). Kenya Health Policy 2014-2030-Towards attaining the highest standard of health. In: health Mo, ed. Nairobi, Kenya: Kenya Ministry of Health; 2014.
[14]	O'Neill K, Takane M, Sheffel A, Abou-Zahr C, Boerma T. Monitoring service delivery for universal health coverage: the Service Availability and Readiness Assessment. Bulletin of the World Health Organization. 2013; 91: 923-931.
[15]	Scharadin BP. Principal component analysis of state level food system Indicators. 2012.
[16]	Kimani L, Namusonge GS. FACTORS INSTRUMENTAL TO SUSTAINABILITY OF PROJECTS IN KENYA: A CASE STUDY OF OPARANYA MOTHER CARE KAKAMEGA COUNTY. Strategic Journal of Business & Change Management. 2016; 3(4).
[17]	HPP. HEALTH POLICY PROJECT/KENYA Building capacity for improved health policy,advocacy, governance, and finance. 2014.
[18]	Koikai JS. An Evaluation of the Effects of Devolution on Healthcare Delivery in Nakuru County 2015.
[19]	AWINO OE. RESPONSE STRATEGIES ADOPTED BY THE MINISTRY OF HEALTH TO CHALLENGES OF DEVOLVED HEALTHCARE SERVICES IN KENYA, SCHOOL OF BUSINESS, UNIVERSITY OF NAIROBI; 2016.
[20]	Miriti AK, Keiyoro P. Influence of devolution of government service delivery on provision of healthcare: A case of level five hospital in Meru County, Kenya. International Academic Journal of Information Sciences and Project Management. 2017; 2(1):316-334.
[21]	Amek N, Vounatsou P, Obonyo B, et al. Using health and demographic surveillance system (HDSS) data to analyze geographical distribution of socio-economic status; an experience from KEMRI/CDC HDSS. Acta tropica. 2015; 144: 24-30.
[22]	Were V, Buff AM, Desai M, et al. Socioeconomic health inequality in malaria indicators in rural western Kenya: evidence from a household malaria survey on burden and care-seeking behaviour. Malaria journal. 2018; 17(1): 166.