S.8. Srikantia
The National Nutrition Monitoring Bureau
Information on prevailing patterns ofdietary intakes and current nutritionalstatus of representative populationgroups in the country is necessary forthe formulation and implementation ofmeaningful national health and nutritionpolicies. In order to derive such information, numerous diet and nutrition surveyshad been carried out in the country during the last few decades. Most of thesesurveys had been undertaken in earlieryears by the nutrition sections of thePublic Health Directorates of differentstates. The results of these studies were
annually compiled and published by theNational Institute of Nutrition (and in earlier years its precursor, the Nutrition Research Laboratories). On the basis ofthe data gathered from these surveys,the National Institute of Nutrition hadalso published a diet atlas and a nutritionatlas both of which had attracted wide
attention on the part of scientists, policymakers and planners in the country.
However, the major weakness in theabove efforts was that the results of sur
veys carried out in different locations bydifferent agencies were not alwaysstrictly comparable. Sampling designswere not strictly standardised. It becameobvious that in order to derive scientifically valid and authentic data, whichwould provide a reliable picture of thenational situation with respect to dietaryintakes and nutritional status, the accuracy of the. data as well as their comparability had to be ensured. It was inrecognition of this that in 1972 the thenDirector of the National Institute of Nutrition (Dr. C. Gopalan) conceived the ideaof setting up the Nation~ NutritionMonitoring Bureau as an integral part of
the National Institute of Nutrition.The Bureau began its activities with
nine peripheral units - one in each ofthe following states - Andhra Pradesh,Gujarat, Karnataka, Kerala, MadhyaPradesh, Maharashtra, Tamil Nadu,Uttar Pradesh and West Bengal. Fiveyears later, Orissa jOined the Bureau asits 10th unit.
The NNMB thus has two clear and
major objectives:• To collect on a continuing basis,
analyse and report on dietary intakesand nutritional status of representativesegments of the population in each ofthe states, employing standardised anduniform procedures and techniques, and
• To periodically evaluate on-goingnational nutrition programmes to identifythe strengths and weaknesses and recommend mid-course corrections whencalled for.
The NNMB has been functioning forwell over a decade (now nearly 16years). An attempt has been made hereto evaluate its contributions.
Some Unique Features Of NNMBData Collection
The representativeness of the samplecovered by the study is ensured by arigorous multistage sampling procedure,by which every village in a state has aneven opportunity to be included. Districtsare stratified into one of four developmental categories and in each of the fourcategories, one third to one sixth thenumber are selected each year for survey. All districts in the state would thusbe covered within a three to six yearperiod, 'depellding upon the size of the
state. The selection of households in
each district is so made that the percen!of rural population covered in each district conforms to the ratio of rural to totalpopulation of that district. Selection 01
villages is made after categorising the'Tinto small, medium and large, dep ,dirupon the population, and then b' ..Jsir,random numbers. Depending ur .. thvillage population, 5, 10 or 20 '0IJ!)eholds are included in the stud~ Th(selection of households is made by If
survey team after it visits the village, S(as to ensure that different socioeconomic and occupational categoriesare represented.
This sampling frame was evolvedafter in-depth discussions with Dr. C.R.Rao - the then Director of the IndianStatistical Institute, Calcutta, and currently National Professor.
In the urban areas, the capital city ofthe state is included, for reasons of convenience of operation, and apart fromthe low, middle and high income groups,one slum and one big industrial concernin the city are covered. Clearly, theseare not representative samples of eitherthe urban, slum or industrial labourpopulations (National Nutrition Monitoring Bureau - Plan of Operation, NIN1971).
Data on food consumption at thehousehold level are obtained through direct weighing of all raw foods used bythe family, on three consecutive days. Atthe individual level information on intakes is obtained by the 24 hour oral recall method. Thus data are collected notonly on total family food consumptionbut also on intra-familial food distribu
tion. This helps in cross-checking thevalidity of the data generated on totalfamily food consumption.
Since data on socio-economic factors,dietary intakes, and nutritional status both anthropometric and clinical - areall obtained from the same households,it would be possible to inter-relate them
in terms of cause and effect.
In each state, all data are collected bya team consisting of a medical officerand a nutritionist, who are speciallytrained at NIN.
Standardised techniques are usedboth for diet surveys and anthropometricmeasurements, thus ensuring uniformityof methodology, to the extent possible. Itneeds to be recognised that despiteusing the best and most appropriatesampling methods, true representativeness can never be ensured. Similarly,the inherent limitations in the best dietsurvey methods pose problems of interpretation and a three-day weighmentmethod or a 24-hour recall method cannot be projected as being truly valid forlonger periods. Give'nthese limitations, itwould be fair to say that the data generated and reported by the NNMB areperhaps the most authentic that are available in our country today. They can beused with confidence not only for interregional comparisons, but also for comparisons across time.
Contributions Of The NNMB
Coverage: Since its inception in1972, till the end of 1984, the Bureauhas covered a total of 39,129 households for diet surveys and 243,536 individuals for assessment for nutritionalstatus, in rural areas. In urban areas corresponding figures are 8,207 and32,500. The Bureau has published itsfindings in a series of reports brought outannually. Data have been presented notonly for the state as a whole, but alsoseparately for each of the districts covered.
Apart from this regular annual datacollection, the Bureau has also carriedout two special studies. In 1983-84, anNNMB-NSSO linked study, to relate theexpenditure-pattern to dietary consumption was done, while in 1985-86 specialtribal groups in the Integrated Tribal Development Project Areas were comprehensively studied. The reports ofthese studies are not as yet available.
Programme evaluation: The NNMBhas evaluated the following nutritionprogrammes and published the findingsas special reports.
• Vitamin A prophylaxisprogramme 1977-78
• Applied NutritionProgramme 1977-78
• India Population Project,Karnataka 1979
• Special NutritionProgramme 1979-80
• World Food-assistedFeeding Programme 1981-82
• National Anaemia
Prophylaxis Programme(Report not yet published) 1986-87
The coverage of both households andindividuals in the Survey looks large andimpressive, as does the fact that everymajor nutrition programme, except theNational Goitre Programme has beensystematically and objectively evaluatedfor its impact, on at least one occasion.The numbers when viewed in the context of the country's vast population maylook small and may give one the impression that the larger numbers shouldhave been covered - particularly in thesmall villages. But considering themeticulous sampling system employed,the figures are not inadequate to drawvalid conclusions.
The original plan of completing datacollection from each state within a periodof three to six years has not beenachieved; nor has the NNMB been ableto revisit a single area which it had earlier covered, to examine directly, thechanges which might have taken placeas a result of the various developmentaland nutrition inputs. Though this activityhad been originally envisaged, it wouldappear that several factors, includingsome administrative difficulties, hadcontributed to this situation.
Scientific Data Generated
It is not the intention here to go into allthe data generated by the NNMB, but tohighlight some of these, which are ofpractical importance. Detailed information may be obtained from the Bureau'sAnnual Reports, published by the NIN.
Intake of food and nutrients and
magnitude of adequacy/inadequacy:(i) Because of the data generated by theNNMB, it is now possible to provide information on the mean intakes of foodand nutrients of households and individuals and also to provide quantitativeestimates of the proportion of familiesand subjects in rural India, whose nutrient needs are not met (mean - 2SD ofthe recommended allowances are usedas cut-off levels). The wide variationsbetween states have become clearer thanbefore as a result of valid comparisonswhich can be made, between regions.
(ii) Data on nutrient intakes havehelped to place on a very firm scientificfooting the observation that the primary
limiting nutrient in the diets of even thepoorest segments of rural India isenergy and not protein - a finding ofenormous practical importance firstmade by Gopalan and co-workers several years ago. This situation is true of allthe states in the country.
(iii) Data on the intrafamilial distribution of food has shown that a higherproportion of preschool children get lessthan their physiological needs, as compared to adults - again a finding common to all the states studied.
(iv) Contrary to the widely prevalentbelief that in situations where a house
hold has insufficient food, the man getsa higher share than does the woman,the NNMB data show that there is no
such discrimination. Proportion of menand women who have calorie inadequacy is similar. Similarly, contrary topopular belief, there is no gender biasagainst girls (Kamala S. Jaya Rao: NFlBulletinVol.5, No.3, 1984).
(v) The NNMB data have made it possible to relate factors such as land holding status, occupation, income and family status (whether joint or nuclear) toadequacy of food and nutrient intake(Nutrition News, NIN Vol. 3, No.2,1982).
(vi) The data have also indicated thatthe measurement of the consumption ofjust one food item - cereals - can beused as a proxy for total energy intake tomake rapid, approximate estimates ofchanging trends in calorie consumption.
Nutritional status: (i) As with foodand nutrient intakes, the NNMB datahave made it possible to provide quantitative estimates of the growth status ofpreschool children - a measure widelyused as a proxy for nutritional status.The types of anthropometric measurements recorded have made it possible toclassify children according to gradesand types of malnutrition, as also tomake inter-state comparisons.
(ii) The simultaneous recording ofspecific clinical nutritional deficiencysigns, has not only provided informationon the prevalence rates of deficiencydiseases, but also to relate growthstatus to clinical deficiency signs.
(iii) Anthropometric measurements ofadult men and women in rural India wasscanty. The NNMB has provided thismuch needed data The widely held im
pression that rural women are more undernourished than are men is not borneout by these data (Kamala S. Jaya Rao,NFl Bulletin Vol. 5, No.3, 1984).
(iv) The expected socio-economic
gradient in nutrient intake has been confirmed. The energy and protein inta.kesof urban slum dwellers are no way dIfferent from those of the rural landlesslabourers (Kamala S. Jaya Rao, N~IBulletin Vol. 6 No.4, 1985). The nutntional status of even the urban middle income groups is far from satisfactory(Kamala S. Jaya Rao, NFl Bulletin Vol.7, No.1, 1986). .
(v) The nutritional status of rural girlsof preschool age has been found to beessentially similar to that of corresponding boys, indicating that there is no gender bias in nutrition care (Gopalan, NFlBulletin Vol. 8 No.4, 1987).
(vi) Data generated by the NNMBhave raised a host of issues related tothe interrelationship between nutrient intake on the one hand and body size onthe other, as well as the importance ofnon-nutritional factors in determiningnutritional status (Gopalan, NFl BulletinVol. 5 ,No.1, 1984; Vol. 6, No.4, 1985).
Programme evaluation: The evaluation reports published by the NNMBhave helped in identifying the correctivesteps which need to be taken, with respect to the Vitamin A ProphylaxesProgramme, the SNP and the WorldFood-assisted Feeding Programme. Italso identified the basic weaknesses inthe Applied Nutrition Programme - theineffective education component.
Changes in the food and nutrilion- situation over time: (i) The Bureauhas found that there is a changing pattern in the types of food grains beingconsumed. The consumption of cereals- wheat and rice - show an upwardtrend, while that of millets, jowar, bajra,ragi, and maize show a downward trend.The mean consumption of vegetables,fruits, milk, oils and fats and sugar haveremained essentially unchanged.
(ii) According to the NNMB, over thelast decade there have been changesboth in the mean calorie intake of ruralIndian households and in the prevalenceof severe forms of growth retardationamong preschool children, at the macrolevel (aggregated data). The Bureau'sconclusion that "during the period from1975 to 1980, there has been an increase in average calorie consumptionlevels of rural populations at the household level" and that "the average intakeof calories in the lowest income groups(less than Re. 1 per person per day)show a definite increasing trend" (Nutrition News, Vol. 7, Nos. 2 and 5, 1986)have come in for sharp criticism (K.
Ramachandran, NFl Bulletin, Vo1.8, No.1, 1987) This criticism has been ~a~edon consideration of both the statisticalmethodology employed and on the observation that the pattern of coverage ofhouseholds according to income .overtime had changed - in favour of higherincome. (The mean increment in calorieintake between 1975 and 1980 has b~enaround 100 calories per consumptionunit.)
The NNMB, however, argues that thevalidity of some of the statistical pointsmade by Ramachandran is debatable. Italso points out that the per capita dataare derived depending mainly on occu
pation and that when intakes of separateoccupational groups are looked at, thetrend of increasing calorie consumptionwould be evident, the largest increasebeing in the landless labour group. However, NNMB's data in the years subsequent to 1980 indicate that the earliertrend of increasing calorie consumptionclaimed by it had been reversed. Thissignificance which the NNMB has attached to marginal variations perhapsneeds to be reassessed.
(iii) The NNMB has indicated that theirdata on the nutritional status of rural Indian preschool children show a generaltendency to improve from 1976 to 1980,based on the observation that the proportion of children with severe grades ofmalnutrition has fallen from 8.5 percentin 1976 to around five percent in lateryears. A look at the earlier NNMB reports shows that there was a suddendrop in this value from over 21 percent in1975 to 8.5 percent in 1976. It needs tobe emphasised that this sensational decline was due to the switching over to adifferent standard of normalcy - from theHarvard to the Hyderabad standard. Thelatter was evolved on the basis of inves
tigation of a relatively small sample ofchildren belonging to well-to-do familiesin Hyderabad. The validity of the use ofthis standard has been questioned onthe basis of the finding that when thisstandard is used, more boys than girls inrural India show greater degrees ofgrowth retardation! This may lead one toan erroneous conclusion that boys,rather than girls are discriminatedagainst (C. Gopalan, NFl Bull. Vo1.8,No.4, 1987). It has also been pointed outthat despite the NNMB's finding aboutthe lower prevalence of severe growthretardation, the mean values for heightsand weights of rural Indian children hadchanged little over the years (Gopalan,NFl Bull. Vol. 7, No.1, 1986).
The need for caution and circumspec-
tion in the interpretation of valuab~edatahardly needs emphasis. It may mdeedbe prudent on the part of NNM.Bin fu~ureto have in-depth in-house discussionson tentative conclusions and have them
critically examined by its own exp.ertgroup before such co~clu~ion~, ~hlchcould carry wide practical Implications,are publicised. This will avoid unnecessary confusion and criticism.
It needs to be emphasised here thatthe criticisms considered above pertainto interpretations of some data; the reliability of the data themselves gathered byNNMB meticulously has not been
questioned. The criticism of interpretations should not be allowed to obscureor belittle the solid achievements. Thepoint to remember is that the data beinggenerated by NNMB are about ~heon!yreliable and authentic data available mthe country today regarding dietary intakes and nutritional status of representative population groups in some parts ofthe country. There can also be no doubtabout the great potential value of thesedata.
Some Limitations Of The NNMB And
Suggestions To Remedy Them
The contributions of the NNMB are
quite impressive. But for it to express itsfull potential and play the important roleexpected of it, it is necessary to correctsome deficiencies.
• An obvious and serious weakness.of the Bureau relates to its incompletecoverage of the country. Its activities arelimited to just 10 states. Some of themost prosperous and developed states- Punjab and Haryana - and none ofthe North-Eastern states are covered,
thus raising the valid objection that current data are not applicable to the country as a whole. This situation needs to bequickly corrected.
• The Bureau's inability to completedata collection within the time frame
originally conceived is also a seriouslimitation. Apart from the data becoming'outdated' to some extent, the validity ofpooling of yearly data to arrive at thestate position may become questionable. The category into which a districtwas placed may change since development is a dynamic process and theBureau may do well to look into whetherthe original indices employed shouldcontinue. To achieve time-bound coverage, expansion of staff and associatedfacilities must be seriously considered.
• Most data-collecting bodies makeuse of but a small fraction of what is col-