Friday, November 19, 2010

Numbers: The Immensity of Undernutrition


Some of us like to think in terms of numbers. It is a way of rationalizing what typically seem to be abstract concepts in the world. Hence, it only makes sense that we seek qualitative evidence to support our claims, especially when gauging the immensity of a situation.

In Public Health, these numbers are used in the very same way. Data is collected, analyzed, interpreted, and disseminated in a process called surveillance. And through surveillance, Public Health agencies can monitor health events and be better prepared to plan and implement different interventions when trends reveal a problem.

The start of the new millennium in 2000 marked the beginning of a new generation of philanthropy. Sustainable development placed a renewed emphasis on the concepts of prevention, which in turn sparked our interests in the trends plotted by surveillance. Just ten years before in 1990, several developing countries faced problems with a high number of undernutritutioned children, who either were not eating nutritionally effective foods or were limited in their access to food in general. With the hope of decreasing these upward trends, the Millennium Development Goals encompassed halving the number of undernutritioned children by 2015. While most countries have comparatively succeeded in this fifteen-year escapade, South-Asia seems to have stagnated in their efforts.

This is particularly the case for India. Compared to global efforts, rates of undernutrition in Indian children under the age of five from 1999 onwards has been decreasing, but too slowly to make any significant difference in the number of consequential health problems that undernutrition brings.

The National Family Health Survey is the primary method of collecting data for Public Health incidence and prevalence studies within the country. However, even though it has released three rounds of data since 1992, it focuses on obtaining data from a representative sample of households, which compromises the accuracy of the estimates. Within the several categories of the survey, recording information about nutrition, access to food, and undernutrition-related disorders (like anemia) has helped track the progress of the Millennium Development Goal [1]. To address the growing concern of nutritional health in 1972, the National Nutrition Monitoring Bureau also began collecting data through a self-reporting survey [2].

Quantitatively, it is quite obvious that India’s progress nutritionally is stunted. In 1990, an estimated 53.3% of the children under the age of three were underweight. By 1998, this number had fallen to around 47% [3]. More recent studies published estimate the number of children under five who are underweight in India to remain around 47% and 46%, with stunting at 45% in 2000 [3]. When this rate was observed again for change in 2005 and 2006, it remained stagnant. It is expected for it to only drop to 40% by 2015, missing the target for the Millennium Development Goal by close to 14% [4].

Yet, the measure of the number of underweight children is not the only indicator of undernutrition. Realizing the stagnation of the rates depends on the comparative data available for different economic brackets of India and other countries. In urban India the percentage dropped from 44% to 38% between 1992 and 1998. In rural India, however, the percentage dropped from 55% to 50%. The rates of change are similar, suggesting that the current interventional method is dissociated with the economy. As of 2000, 24% of African children under the age of five suffered from underweightedness. Latin America, the same year, reported that only 6% of the population of the children under the age of five were underweight. These countries are all exhibit a similar pattern of economic development; in fact, India may even be growing at a faster rate than others in the developing world [3].

This is where indirect indicators play a crucial role in exposing the magnitude of the problem. India’s Gross Domestic Product (GDP) information allows us to determine the comparative effect of the country’s economy in relation to the health needs provided. Yet, by looking at indirect indicators like vitamin deficiency and iron anemia, we can gauge what type of effect undernutrition has on the population. According to recent data, 74% percent of the population of 6 month-olds to 36 month-olds have some form of Iron Deficiency Anemia (IDA) [3].

However, the fundamental problem with all the data is that it is from a survey-based study, which are self-reported. This compromises the accuracy in the numbers, as they are only estimates of the prevalence and incidence.

It seems like numbers define parts of our lives. But, when they can be used to help us create an intervention in Public Health, why not when surveillance can save lives?


Sources: 
[1] http://www.nfhsindia.org/
[2] http://www.nnmbindia.org/aboutus.html
[3] MDG Report 2007
[4] MDG Report India 2009

1 comment:

  1. Well done Kavya. One question to consider: assuming, as you say, the data was collected using a REPRESENTATIVE sample of the families in India, why would that(as you porport)"compromize the accuracy of the estimates"? By definition, if a sample is representative and does not suffer from other sources of bias associated with survey-based sampling, then you can make rather accurate inferences about the population you are studying (target population). So, is the sample representative or not? If yes, are there any other biases associated with this type of data collection present here? What biases can compromize the estimates derived from a self-reported survey (look up information bias, recall bias, etc). Your blog was relatively well organized and well written. However, you quoted lots of data (?maybe too much) and offered relatively short and undetailed analysis of that data. What is the bottom line here? What picture does the data paint? What are the underlying trends and determinants of what the data indicates? What are the sources of uncertainity in the data? What can be done to get better estimates? If you make sound arguments based on available data and knowledge in response to these questions then you will have a great paper.

    ReplyDelete