Malnutrition Analysis 1 
Last Updated: December 17, 2019; First Released: June 10, 2014
Author: Kevin Boyle, President, DevTreks (1*)
Version: DevTreks 2.2.0
A. Introduction
This reference explains how to start to collect, measure, and analyze malnutrition data (2*). DevTreks believes that all malnutrition data, from the nutritional status of Brooklyn kids to Nepalese farmworkers, has stories to tell and lessons to teach. Those lessons can only be learned when data about malnutrition is collected, measured, aggregated, analyzed, explained, and saved in online knowledge banks. Full, uniform, and accurate analyses of the nutritional status of Bangladeshi children, Appalachian seniors, California farmworkers, and Nicaraguan factory workers, should be one or two links away for everyone. If a malnutrition expert, business owner, parent, government official, or nonprofit member, needs to make a decision involving malnutrition, they should have ready access to the best data and advice available.  This reference introduces another DevTreks way to build these types of knowledge banks.
Section
Page
Background, Calculations, and Data URLs
2
Work Breakdown Structures and Rules
3
Multipliers
6
Analyzers
6
Performance Analysis
9
Sustainable Food System Supply Chain Analysis
9
Multimedia and Stories
15
Knowledge Banks and Summary and Conclusion
17
Appendix A. Multipliers
20
Appendix B. Malnutrition Analysis
27
A. Background, Calculations, and Data URLs (4*)
The Malnutrition Calculation 1 reference explains the food nutrient calculations that will be analyzed in this reference. Appendix A in that reference introduces the background logic for conducting these studies. One of the initial food nutrition Input data sets consists of 7290 Inputs, each containing a food nutrition calculator that holds their USDA, Agricultural Research Service (ARS) Standard Reference (SR) nutrient composition. The localhost database includes two ARS food nutrition Input data sets. Only the SR Inputs have calculators attached (i.e. the FNDDS5 does not). They can be distinguished from the non-SR by their capitalized name. The introductory reference also explains how to manage nutrient data sets (i.e. bulk uploads).
All of the analysis displayed in this reference will be proforma sample data that demonstrates how to collect and analyze each type of dataset. This reference used the localhost (Version 1.6.5) and cloud (Version 2.1.0) deployments to document calculations. The video tutorial also uses Version 2.2.0 because of further advancements in code and tutorials. This data belongs to the Family Budgeting and Food Nutrition club (if necessary, switch default clubs).
The Malnutrition Analyzers demonstrated in this reference can be found at:
https://www.devtreks.org/hometreks/preview/smallholders/linkedviewgroup/Food Nutrition Analyzers Group/21/none
Sample data can be found at: 
https://www.devtreks.org/hometreks/preview/farmworkers/input/BARLEY,PEARLED,RAW/2147395842/none/
https://www.devtreks.org/hometreks/preview/smallholders/output/BARLEY, PEARLED, RAW/2141211289/none/
https://www.devtreks.org/hometreks/select/farmworkers/outcomegroup/Food Subsistence Meals/38/none/
https://localhost:5001/hometreks/preview/farmworkers/outcomegroup/Crop Foods/41/none
https://www.devtreks.org/hometreks/preview/smallholders/operationgroup/Food Basic Stock Operations/759/none/
https://www.devtreks.org/hometreks/preview/smallholders/componentgroup/Food Subsistence Supplies/533/none/
https://www.devtreks.org/hometreks/preview/smallholders/budget/Food Nutrition Subsistence Stocks SR01/273083905/none
https://localhost:5001/hometreks/preview/farmworkers/budgetgroup/Food Nutrition, SR Budget Analyses/2140761977/none
https://www.devtreks.org/hometreks/preview/smallholders/investment/Food Supply Stock Budget/390/none
https://www.devtreks.org/hometreks/preview/smallholders/resourcegroup/Food Nutrition Tutorial/134/none/
https://www.devtreks.org/hometreks/preview/smallholders/linkedviewpack/Food Nutrition Tutorial 1/170/none
With the exception of Input and Output Analyses, NPV calculators must be run prior to running these analyzers. DevTreks convention is to use NPV calculators to pull fresh database data together prior to running analyses and to conduct basic benefit cost analysis. The NPV calculators do not rerun food nutrient calculations, but all of the analyzers explained in this reference do. The NPV calculators document the benefits and costs of the combinations of Inputs and Outputs (i.e. the cost of a localized meal).
B. Work Breakdown Structure (WBS) and Rules
Food Input data used the USDA, Agricultural Research Service Standard Reference (SR) WBS. The following image demonstrates how the SR WBS is used to structure Input data used in this reference. Although not shown in this image, we recommend including the year (2014 ACORN) in the names of Inputs, Outputs, Operations, Components, Outcomes, and Time Periods.

No WBS was found for Outputs, Outcomes, Operations, Components, Operating Budgets or Capital Budgets. The Ag Production Analysis 1, Construction Analysis 1, and Health Care Analysis 1 tutorials demonstrate how to structure full data sets to support the analyses shown in this reference. With the exception of Inputs, the actual data used in this analysis was structured for the purpose of testing these analyzers. 
C. Multipliers
Appendix A. Multipliers, documents how base element multipliers (i.e. Input.Times, Output.CompositionAmount, Outcome.Amount …), can be used to adjust the aggregated unit food nutrition calculations contained in Inputs and Outputs. 
D. Analyzers
Malnutrition Totals, Statistics, Change, and Progress, Analyzers are available for examining Inputs, Outputs, Operations, Components, Outcomes, Operating Budgets, and Capital Budgets. Standard aggregators (Label, Group Id, and Type Id) can be used to aggregate the data being analyzed (see the Calculators and Analyzers tutorial). The following image shows that, with the exception of Totals Analyses, up to 10 nutrient properties can be chosen to analyze. 

The following image makes the point that the 10 nutrient properties being studied should be consistent throughout the analyses being conducted. Source code users can change the number of nutrients being studied to any amount desired, but keep in mind the limitations of displaying html data. The number of observations used in any analysis will reflect the number of base elements being aggregated.

Appendix B. Analysis, explains how to use the Totals, Statistics, Incremental Change, and Progress, analyzers.
E. Performance Analysis
The data generated by analyzers can be used to carry out other types of Performance Analysis, such as input per unit output, output per unit input, and cost per unit output (see the Performance Analysis 1 reference). The IPCC WG1 2013 reference describes how Net Primary Productivity crop studies, that make use of water and nutrient budgeting techniques, will play increasingly important roles in future efforts to adapt to climate change impacts on crop productivity (and decreased pollution from runoff) (5*). 
F. Sustainable Food System Supply Chain Analysis (6*)
Version 2.1.0+ included a Social Performance Analysis tutorial demonstrating how to assess the sustainability of food systems. The first 2 paragraphs were taken from the Social Performance Analysis 2 reference. The 3rd paragraph was adapted from the Health Care Analysis reference.
Notarnicola et al (2017) review the state of LCA art in food system analysis and highlight the numerous challenges remaining to be tackled in this industry. These challenges include 
* the need for dietary shifts to sustainable food systems 
* field level LCAs that don’t adequately address landscape level sustainability impacts on soil quality and fertility, land erosion, reduced ecosystem services, and biodiversity loss 
* integration of social, economic, and cultural factors into LCA studies 
* the reliance on average LCAs for predominant food production systems rather than the reality of extreme production variability 
* technical deficiencies dealing with product quality, geographical contexts, temporal variability, machinery, functional units, ecosystem services and biodiversity 
* consumer education that results in behavioral change 
* missing supply chain phases such as food waste 
* integration with “mixed methods explanatory approaches” used in food production studies 
* accounting for accidents and disasters
Notarnicola et al (2017) point out that failure to address these challenges could surpass planetary food system sustainability, and therefore food security, needs. In the context of this reference, it’s the job of social networks and clubs to address those challenges, including the development of open access, rather than commercial, sustainable food system databases that make all of their TEXT data available through URIs.
Even if the algorithms and techniques introduced in these tutorials were perfect, the food industry in many countries may still not deliver sustainable food system services better. Perfect software, technology, or algorithms, won’t make any difference when the real issues involve institutional failure (8*). This reference recommends consequential digital activism that allows consumers and “good actor” food industry professionals to independently assess the performance and accountability of all parts of the food system supply chain, from research to consumer consumption. The direct consequences need to lead to better decisions about purchases, penalties, punishments, incentives, and redistributions.
For example, Poore et al (2019) built the following international food supply chain database to demonstrate tying mitigation policies to agricultural externalities that include GHG emissions, eutrophying emissions, acidifying emissions, energy use, and freshwater withdrawals.  The authors use the following statement to confirm their assessment of the sustainability of the whole supply chain:
 “The system we assess begins with inputs (the initial effect of producer choice) and ends at retail (the point of consumer choice).”

Rose et al (2019) use the following table to demonstrate how to use U.S.A food dietary databases to link the carbon footprint of consumer diets to consumer demographic and behavioral factors (i.e. SPA3’s Social Impact Analysis). The authors use the following statement to confirm their focus on sustainable consumer diet choices:
“to assess the GHGE from individual self-selected diets in the United States and examine their association with nutritional quality of the diets, demographic patterns, and food-related behaviors.”

In keeping with the sustainability spirit of both studies, Version 2.2.0 investigated whether the “Extra” food nutrient calculator properties could be used to study simple “carbon footprints” (i.e. GHG emission and Energy Use). The following food nutrient Operating Budget and Emission Balance Budget confirms that these properties support the calculation of final environmental footprint balances. Although these numbers are fictitious, databases similar to the 2 previous datasets can be used to set these properties to actual amounts.


https://localhost:5001/hometreks/preview/farmworkers/budget/Nutrition Budget 01 Benchmark/273083912/none
 

Poore et al (2019) recommend using this type of approach to help producers identify and adopt sustainable mitigation practices and to assist consumers make sustainable purchases. Rose et al. (2018) recommend using this type of approach to help targeted consumers eat more sustainable diets. These new datasets and examples demonstrate that consequential digital activism is not only possible –it’s under way (refer to Footnotes 7 and 8 in the associated Calculation reference).
G. Multimedia (Resources)
All analysis should be accompanied by multimedia that help to explain the malnutrition intervention. The multimedia can include graphs and other visual aids that help users to interpret all of the data. The economic and nutritional characteristics of the following types of agricultural production, in this case a mixed vegetable crop, are easier to interpret with fuller multimedia support.

H. Stories (Linked Views)
All analysis should be accompanied by stories that explain the malnutrition intervention. 


I. Knowledge Bank Standards
All malnutrition data should be entered into online knowledge banks (i.e. production servers as contrasted to development servers) that can be used to analyze malnutrition. That structured evidence must be passed down to future generations. These knowledge banks aggregate and analyze all of the data in a network. Future references will discuss how these knowledge banks will evolve (i.e. supply chain sustainability forecasts) to support future decision making needs. The flexibility offered by DevTreks in documenting malnutrition means that networks need to develop “rules” explaining the “standards” that should be followed by clubs in their network. The “standards” make it possible to build knowledge banks. 
Summary
Clubs using DevTreks can start to carry out the basic analysis of malnutrition. Clubs can solicit help understanding malnutrition better and share structured evidence explaining malnutrition. Networks can build knowledge banks that explain malnutrition and pass that knowledge down to future generations. The result may be Congolese smallholders who raise healthier children, Pacific Islander adults who tackle obesity more effectively, inner city school administrators who deliver better malnutrition programs, consumers who only purchase sustainable products, producers who ensure the sustainability of their full supply chains, investors who only invest in sustainable companies, and people who improve the sustainability of their lives and livelihoods.
Footnotes
1. Although the author has studied malnutrition as an important, but ancillary topic, in his agricultural science education at Cornell University, USA (B.S.) and U.C. Davis, USA, (M.S.), he is not an expert in the field. The tools introduced in this reference were kept basic for that reason. 
2. In the course of building the tools introduced in this reference, several additional tools were investigated. The most promising tracked household food nutrient consumption and production. The Malnutrition Calculation reference discusses domain-specific software patterns versus alternative designs in greater depth.
3. Nutrient stock budgets are commonly used to study soil and plant nutrient balances for managing crop nutrients. Refer to Footnote 2. 
4. A small, limited amount of nutritional data was used to test the analyzers in this reference. In addition, not every feasible way to run an analysis was tested. These analyzers will continue to be tested with additional data sets in future upgrades. 
5. Although these types of studies are becoming more important, researchers have been conducting them for years –the author spent the summer of 1978 working on a Cornell University research farm experiment involving corn yield and C02 use.
6. Monetary compensation for software development has never been a priority for DevTreks. We believe malnutrition is a public goods problem that is best tackled with public goods software. But we recognize that the private sector is better at attracting the resources that are needed to tackle these types of global problems. We encourage software companies to investigate business models that provide them with fair compensation while still addressing public goods problems (i.e. unless their primary motivations involve money, greed, self-interest, or narrow-mindedness).
References
The references used in this tutorial can be found in the Malnutrition Calculation 1 reference.
References Note
We try to use references that are open access or that do not charge fees.
Improvements, Errors, and New Features
Please notify DevTreks (devtrekkers@gmail.com) if you find errors in these references. Also please let us know about suggested improvements or recommended new features.
A video tutorial explaining this reference can be found at:
https://www.devtreks.org/commontreks/preview/commons/resourcepack/Malnutrition Analysis 1/450/none/


Appendix A. Multipliers
The following examples use an Operating Budget Analysis of a Turkey slice Input to explain how multipliers work. All multipliers come from the before-aggregated elements. The blue motif reflects 2014 documentation.
The following image displays the initial Turkey Breast Input example explained in the Malnutrition Calculation 1. No multipliers have been used in this analysis.
Version 2.1.0 retested the Outcomes to verify that the multipliers work.

The following image shows that when the Input.OCAmount property is changed from 1 to 2, the food nutrients, serving size, and serving cost properties double.
 
The following image shows that when the Input.Times property is changed from 1 to 2, all of the Input and Operation properties double (Output.Times work similarly).
 
The following image shows that when the Operation.Amount property is changed from 1 to 2, all of the Input and Operation properties double (Outcome/Component.Amounts work similarly).

The following before and after images show that when the TimePeriod.Amount property is changed from 1 to 2, the Input and Operation properties don’t change but the TimePeriod properties all double. Again, the before-aggregated TimePeriod multipliers are used. 

After





Appendix B. Malnutrition Analysis
This Appendix explains how to conduct Totals, Statistics, Incremental Change, and Progress, analyses of the food nutrition Input and Output calculations documented in the sibling Malnutrition Calculation reference. The following images come from both localhost:5001 and cloud datasets. No major attempt is made to keep the 2 datasets fully consistent with one another.
a. Totals Analyses
A Totals Analysis sums all nutrient properties for every base element in an analysis. All analyzers run this analysis for each aggregated base element before carrying out additional calculations. All totals derive from initial malnutrition calculations. Malnutrition calculations are run prior to summing the calculations. 
Operating and Capital Budgets will subtract Outcome sums from Operation/Component sums. The subtraction can serve as a simplified nutrient stock budget where the amount of nutrients entering a system is offset by the amount of nutrients leaving a system (3*). In this type of nutrient stock budgeting, Inputs are credits to the system while Outputs are debits to the system.  For example, a common definition of calorie balance (USDA, 2010) is:  “The balance between the calories consumed in food and the calories expended through physical activity and metabolic processes”.
Operating Budgets use the techniques explained for analyzing food production and consumption in the Malnutrition Calculation 1 reference (see the examples for Food Consumed Nutrient Content and Food Produced or Expended Nutrient Content). If food production Outputs are left out of an Operating Budget, the result will be a summation of all the food nutrients consumed in Inputs. If food consumption Inputs are left out of a budget, the result will be a summation of all the food nutrients produced or expended in Outputs (but they’re still subtracted from Inputs with a zero amount, resulting in negative numbers).
Capital Budgets use the techniques explained for analyzing food supplies in the Malnutrition Calculation 1 reference (see the examples for Food Supply Container Nutrient Content and Food Distribution Container Nutrient Content). If food supply Outputs are left out of Capital Budgets, the result will be a summation of all the food nutrients supplied in Inputs. If food supply Inputs are left out of a budget, the result will be a summation of all the food nutrients distributed in Outputs.
The following Operation Totals Analysis shows that summations of Inputs and Outputs with different units of measure will have good food nutritional totals and Serving Costs but not necessarily Container Size, USDA Servings per Container, Serving Size Units, or Serving Sizes. The individual Inputs and Outputs still have legitimate values for those properties.

The following Operating Budget Analysis demonstrates that although the Outcome and Output have positive nutritional summations, the Time Period has some negative numbers because the nutritional Outcome totals are being subtracted from the Operation totals.  

The following Operations are children of the previous image’s Time Period. The TimePeriod.Carbohydrate property is calculated as follows: -71.43 (TimePeriod.Carbo) = 87.2 (Operation.Carbo) – 158.6 (Outcome.Carbo).

b. Statistics Analyses
A Statistics Analysis uses the Totals calculations to measure basic statistical properties of up to 10 aggregated food nutrition properties. Total, Median, Mean, Variance, and Standard Deviation statistics are generated for all of the nutrient properties in aggregated base elements. Nutrients are aggregated in two stages. The first stage uses the standard aggregators to aggregate the base elements. The second stage aggregates the same nutrients within each aggregated base element. The number of observations reflects the number of base elements being aggregated. 
The following three images derive from a three year Operating Budget Statistical Analysis. They demonstrate that the nutritional composition of agricultural Outputs and food Inputs can be included in an analysis. The basic nutritional stock budgeting (Inputs minus Outputs equal current year nutrient stock contributions) is a good foundation for additional types of nutrition decision support tools (3*).





The following three year Outcome Statistical Analysis examines a potential crop output’s nutritional properties. Outcomes can also be combinations of food nutrient Outputs that are expended (i.e. by physical activities and metabolic processes).

c. Change Analyses
Change Analyzers can examine incremental changes in 10 food nutrition properties. A Change by Year Analysis measures incremental changes between aggregated nutrient properties that have different Years. A Change by Id Analysis measures incremental changes between nutrient properties that have different Ids. A Change by Alternative Type Analysis measures incremental changes between aggregated nutrient properties that have different Alternative Types. Changes are analyzed in ascending order (Id = 1,2,3; Year = 2000, 2001, 2002; AlternativeType = A, B, C). The first member of the sequence will be used as a “Base” element to make comparisons. The sibling sequence member immediately before the current sequence member will be used as an “x-1” (x minus 1) element to make comparisons. Gaps in the sequence, such as a missing Year, will be ignored.
The “Base” comparator’s nutrient properties can be adjusted in Inputs or Outputs to targeted goals (i.e. USDA 2010 Nutritional Goals), resulting in analytic goal comparison. Further details about how Change Analyzers work can be found in the Change Analysis tutorial. 
The following three year Input Change by Year Analysis tracks annual changes in food prices, serving costs, and nutrient properties. Under what circumstance might nutrient properties change?

The following proforma Capital Budget Change by Year Analysis measures incremental changes in serving costs, and nutrient characteristics for three years of budgets. The Totals section mentions that Capital Budgets can be used to analyze food supplies, while Operating Budgets can be used to analyze food production and consumption.

The following Operating Budget Change by Alternative Analysis measures incremental changes in food prices, serving costs, and nutrient characteristics for three alternative budgets. What may be the significance of having this type of data for every meal being produced in the world (or in your local house)?

d. Progress 1 Analyses
A Progress 1 Analysis uses the Totals calculations to measure actual versus planned progress for up to 10 aggregated food nutrition properties. Base elements that have a Target Type property set to “benchmark” act as a comparator for base elements using a Target Type property set to “actual”. The “benchmark” comparators displayed below are also used to test the other analyzers and do not have typical benchmark amounts.

The “Benchmark” comparator’s nutrient properties can be adjusted in Inputs or Outputs to targeted goals (i.e. USDA 2010 Nutritional Goals), resulting in analytic target comparison. Further details about how Progress Analyzers work can be found in the Earned Value Management Analysis tutorial. 
The following three year Input Analysis shows that the benchmark comparator is a 2011 Input. Both the 2012 and 2013 actual Inputs are being compared to the 2011 comparator. 

The following three year Output Analysis shows that the benchmark comparator is a 2011 Output. Both the 2012 and 2013 actual Outputs are being compared to the 2011 comparator. This view uses the “compare only” option.

The following three year Outcome Analysis shows that the benchmark comparator is a 2011 Outcome. Both the 2012 and 2013 actual Outcomes are being compared to the 2011 comparator. What may be the significance of having this type of data available for every combination of crops being grown in the world (or at your local organic farmers market)?

The following three year Operation Analysis compares a 2011 Operation benchmark comparator to 2012 and 2013 actual Operations. What might be changing in this diet to cause this type of progress?

The following three year Component Analysis compare a 2011 benchmark Component to 2012 and 2013 actual Components. Components can be analyzed using the techniques explained for analyzing food supplies in the Malnutrition Calculation 1 reference (see the Food Container Nutrient Content Example).

The following three year Capital Budget Analysis shows that the benchmark comparator is a 2011 Capital Budget. Both the 2012 and 2013 actual Budgets are being compared to the 2011 comparator. All three Time Periods are being compared because they have the same Labels and the 2011 Time Period has a Target Type = “benchmark” while 2012 and 2013 have Target Types = “actual”.

The following Operating Budget Analysis compares progress using a benchmark Budget 1 with 3 Time Periods and two actual Budgets, each with 3 Time Periods. At the Budget Level of this analysis, the cumulative progress of actual Budgets 2 and 3 are displayed.
 
The Time Periods in the analysis line up because of the way both Target Type and Label properties were set. At the Time Period Level of this analysis, actual Budget 3’s Time Periods are being compared to benchmark Budget 1’s Time Periods (2011 Time Period Labels = FN2011, 2012 Time Period Labels = FN2012 …). Actual Budget 2’s Time Periods don’t factor in these displayed numbers. Note Budget 2 and 3 use many of the exact same Operations and Outcomes.

The Outcomes and Operations in the analysis line up because of the way both Target Type and Label properties were set. At the Outcome and Operations Level of this analysis, actual Budget 3’s Outcomes are being compared to benchmark Budget 1’s Outcomes (Budget 1, 2011Time Period OC 1 Label = SR01, Budget 3, 2012 Time Period OC 1 Label = SR01 …).  As with Time Periods, Actual Budget 2’s Outcomes and Operations don’t factor in these displayed numbers. Note Budget 2 and 3 use many of the exact same Operations and Outcomes.



  DevTreks –social budgeting that improves lives and livelihoods

1