Article A10.1 Relevance of Secondary Data for Health Policy Research in India Feroz Ikbal Department of Hospital Management Deccan School of Hospital Management, Hyderabad, India ferozikbal[at]yahoo.co.in Published Online: February 18, 2013 1. Overemphasis on Primary Data There appears to be an obsession among Indian researchers for using primary data in health policy studies. This obsession is seen in academic institutions, particularly in doctoral education. Many individual researchers and supervisory committees highlight the importance of primary data for doctoral research, without sufficient justification. For many academic institutions in India, use of primary data is considered essential in a doctoral thesis. This article questions such overemphasis on primary data and argues in favour of secondary data in health policy research. Although direct experience of collecting data from the field could be a valuable part of doctoral education, the health policy researcher also needs to develop other competencies demanded by the sector. The argument presented here suggests a need for reviewing the assumptions behind the typical emphasis on primary data in doctoral studies on health policy. Policy research is critically important for India in the fields of health and education as the country lags behind in these areas in comparison to countries of similar or poorer economic status. Admittedly, the recent years have seen some improvement in healthcare indicators. Some of the credit for this ought to go to the health policy research undertaken by academic institutions, voluntary organisations, and independent researchers. Health policy research in a country like India requires large sample sizes, cutting across geographic, demographic, and other relevant population categories. Obtaining such data directly from the field proves time-consuming and expensive for individual researchers. In comparison, available secondary data constitute a low-cost alternative for policy studies (Best, 1999). Young and Ryu (2000) have described the importance of utilising existing data from both governmental and nongovernmental sources in order to provide timely and credible inputs into the policy-making process. 2. Secondary Data for Policy Research Large secondary datasets are available for health policy research. A prominent source is the Monitoring and Evaluation to Assess and Use Results: Demographic and Health Surveys (MEASURE DHS) project, implemented by ICF International (http://www.measuredhs.com/). The MEASURE DHS project has collected, analysed, and disseminated data on population, health, and nutrition through more than 300 surveys in over 90 developing countries. The project is funded by the United States Agency for International Development (USAID), with contributions from various other national and international agencies. The MEASURE DHS is comparable to the Multiple Indicator Cluster Surveys (MICS) conducted by UNICEF, which is another source of secondary data on the situation of children and women. MEASURE DHS also covers biomarker data relating to health conditions (such as anaemia and HIV infection). Biomarker data are of great importance in health policy research. A biomarker such as haemoglobin level can show the extent of malnutrition. Such data will be prohibitively expensive for individual researchers to collect, even if limited to sample size. 3. Secondary Data Pertaining to Health Status in India National Family Health Surveys (NFHS) and District Level Health Surveys (DLHS) are two large databases which can be used for health policy research in India. NFHS is perhaps the world’s largest health survey. Three rounds of survey have been completed under NFHS (NFHS-1: 1992-1993, NFHS-2: 1998-1999, and NFHS-3: 2005-2006, with roughly a 5-year gap between successive surveys). Around 110,000 women and men were tested for HIV and more than 200,000 adults and young children were tested for anaemia. Commenting on the use of NHFS data in health policy, James and Irudaya Rajan (2008) argue that: (a) NFHS-3 data were not critically evaluated in the policy process, (b) data pertaining to nutrition, immunisation, and gender violence might be of doubtful quality, and (c) data pertaining to infant mortality rate and fertility could be of good quality. Several researchers working on public health in India have used NFHS data in their doctoral theses. Gunasekaran (2008) used NFHS-1 data in his PhD thesis to find out determinants of infant and child mortality in rural India. Wainwright, (2006) used NFHS-2 data to understand gender issues in the family formation process in Uttar Pradesh. She used logistic regression as a data analysis tool. Ikbal (2011) has used NFHS-3 as one of the sources of data in his doctoral thesis to evaluate the National Health Policy, 2002. NFHS-3 data have been used by Mishra (2010) to assess inequalities in childhood nutrition in India. Another secondary dataset which can be used for health policy research in India is the National Sample Survey Organisation (NSSO). NSSO has been collecting data on various dimensions of the Indian economy and society through nationwide sample surveys to assist in socioeconomic planning and policy-making. A survey on “Morbidity and Health Care” was undertaken by the NSSO during 2004, which has been used widely in health policy studies. The Ministry of Health and Family Welfare, Government of India brings out the Bulletin of Rural Health Statistics, which provides data on rural health infrastructure and healthcare personnel across various States and Union Territories. Of course, the data on some states are relatively less comprehensive. 4. Conclusion It is said, “If you can measure it, you can improve it.” India lags behind in several of the healthcare indicators. Some of the States, such as Kerala and Tamil Nadu have performed better and their performance is comparable to that of of middle-income countries. Whereas the performance of some other States, such as Uttar Pradesh and Bihar is poor, with results comparable to the grim health situation of Sub-Saharan Africa. Data obtained from NFHS and DLHS will be of help to policy makers and administrators. Since these two datasets are based on large-scale surveys, researchers can use various statistical tools, such as chi-square test, correlation, ANOVA, regression, and logistic regression. Data from NSSO and the Bulletin of Rural Health Statistics can also be subjected to statistical analysis (e.g., trend analysis). Availability of large-volume data is vital for evidence-based policy making for a country of such diversity as India. Health policy researchers need to know the various secondary data sources available and their relative strengths and weaknesses. In order to play a useful role in the policy process, health policy researchers need to develop the skills of drawing out policy insights from such secondary data. Acknowledgement Thanks are due to Professor D. P. Dash whose comments and suggestions helped me prepare the article for Research World. References Best, A. E. (1999). Secondary data bases and their use in outcomes research: A review of the Area Resource File and the Healthcare Cost and Utilization project. Journal of Medical Systems, 23(3), 175-181. Ikbal, F. (2011). National Health Policy, 2002: An evaluative study. Unpublished doctoral dissertation, Osmania University, Hyderabad, India. James, K. S., & Irudaya Rajan, S. (2008, November 29). Third national family health survey in India: Issues, problems and prospects. Economic and Political Weekly, XLIII(48), 33-38. Gunasekaran, S. (1998). Determinants of infant and child mortality in rural India. Unpublished doctoral dissertation, Sri Venkateswara University, Tirupati, India. Mishra, R. N. (2010). Undernutrition in India: Dimensions and correlates. Unpublished doctoral dissertation, Centre for Development Studies, Thiruvananthapuram, India. (Awarded by Jawaharlal Nehru University, New Delhi, India) Wainwright, S. C. (2006). Gender and family formation in Uttar Pradesh, India. Unpublished doctoral dissertation, The University of Edinburgh, UK. Retrieved from http://hdl.handle.net/1842/1512 Young, W. B., & Ryu, H. (2000). Secondary data for policy studies: Benefits and challenges. Policy, Politics, & Nursing Practice, 1(4), 302-307. Copyleft Xavier Institute of Management, Xavier Square, Bhubaneswar 751013, India Research World (ISSN 0974-2379) http://www1.ximb.ac.in/RW.nsf/pages/Home | ||
 |