========
pygrowup
========

pygrowup calculates z-scores for the following anthropometric indicators:

* weight-for-age

* length/height-for-age

* weight-for-length/height

* head-circumference-for-age

* body-mass-index-for-age

based on the WHO Child Growth Standards:
* http://www.who.int/childgrowth/standards/en/
* http://www.who.int/entity/childgrowth/standards/technical_report/en/index.html

and can optionally use CDC growth standards:
* http://www.cdc.gov/growthcharts

pygrowup avoids floating-point operations to eliminate the unwanted rounding
that muddles the precision of some of the igrowup implementations:
* http://docs.sun.com/source/806-3568/ncg_goldberg.html


REQUIREMENTS
============

* Python 2.5 or later
* Additionally, Python 2.5 requires installation of http://pypi.python.org/pypi/simplejson


INSTALLATION
============
`pip install pygrowup`


EXAMPLE USAGE
=============

Typical usage might look like this::

    #!/usr/bin/env python

    from pygrowup import Calculator
    # helpers contains optional utilities for formatting dates, etc
    from pygrowup import helpers

    # Height adjustments are part of the WHO specification (see section 5.1)
    # to correct for recumbent vs standing measurements,
    # but none of the existing software seems to implement this.
    # default is false so values are closer to those produced
    # by igrowup software
    #
    # WHO specs include adjustments (see Chapter 7) to z-scores of weight-based
    # indicators that are greater than +/- 3 SDs. These adjustments
    # correct for right skewness and avoid making assumptions about
    # the distribution of data beyond the limits of the observed values.
    #
    # However, when calculating z-scores in a live data collection
    # situation, z-scores greater than +/- 3 SDs are likely to indicate
    # data entry or anthropometric measurement errors and should not
    # be adjusted. Instead, these large z-scores should be used to
    # identify poor data quality and/or entry errors.
    # These z-score adjustments are appropriate only when there
    # is confidence in data quality.
    #
    # In this example, Calculator is initialized with its default values
    # (i.e., ``calculator = Calculator()`` would do the same thing).
    # The ``include_cdc`` option will enable CDC measurements for children >5 years.
    calculator = Calculator(adjust_height_data=False, adjust_weight_scores=False,
                           include_cdc=False, logger_name='pygrowup',
                           log_level='INFO')

    # for a timeless example, lets pick a birthdate nine months ago
    import datetime
    great_day = datetime.datetime.utcnow().date()
    nine_months_ago = great_day - datetime.timedelta(days=(9 * 30.4374))

    # nine months ago in an odd, ambiguous string format
    dob = nine_months_ago.strftime("%d%m%y")

    my_child = {'date_of_birth' : dob, 'sex' : 'male', 'weight' : '8.0', 'height' : '69.5'}

    # optionally use helper functions for formatting data

    # transform something like '100309' into '2009-03-10'
    valid_date = helpers.get_good_date(my_child['date_of_birth'])

    # transform 'male' into 'M'
    valid_gender = helpers.get_good_sex('male')

    # calculate 9 months from valid_date
    valid_age = helpers.date_to_age_in_months(valid_date[1])


    # calculate length/height-for-age zscore
    lhfa_zscore_for_my_child = calculator.lhfa(my_child['height'], valid_age, valid_gender)

    # calculate weight-for-age zscore
    wfa_zscore_for_my_child = calculator.wfa(my_child['weight'], valid_age, valid_gender)

    # calculate weight-for-length zscore
    # optional height parameter is only necessary for weight-for-height
    # and weight-for-length
    wfl_zscore_for_my_child = calculator.wfl(my_child['weight'], valid_age, valid_gender, my_child['height'])

    # Note: for backwards compatibility you may still make calls to:
    wfl_zscore_for_my_child = calculator.zscore_for_measurement('wfl', my_child['weight'], valid_age, valid_gender, my_child['height'])


EXCEPTIONS
==========

caller should watch for:

* `AssertionError` raised when caller provides inappropriate parameters

as well as more specific errors (all subclasses of `RuntimeError`):

* `InvalidMeasurement` raised when measurement is invalid for requested indicator

* `InvalidAge` raised when age is invalid for requested indicator

* `DataNotFound` raised when WHO/CDC data is not found for the requested observation (e.g., box-cox, median, coeffeciant of vairance for age)

* `DataError` raised when an error occurs while loading WHO/CDC data into memory


TESTING
=======

install nose to execute tests:
`pip install nose`

the included tests use example anthropometric data taken from
demonstration data shipped with WHO's igrowup software.
pygrowup performs the same calculations and compares the results
to the WHO results.
please see the sofware licence agreement for WHO's igrowup, which
is the souce of the test data files:
http://www.who.int/childgrowth/software/license2.pdf

currently, 4 cases fail to produce results within 1 standard deviation
of the WHO resuts. I believe these discrepencies are due to WHO's use
of floating point arithmetic in their igrowup software, which leads to less
precise calculations compared to pygrowup. In the absence of any other
trusted test data, please be aware that no claims are made to the
accuracy or reliability of pygroup's calculations.

to run the tests:
`$ nosetests tests.py`


DEVELOPING
==========

The source WHO .txt tables can be easily converted to json with the help of
two amazing python utilities:

* The Pyed Piper https://code.google.com/p/pyp/

* csvkit http://pypi.python.org/pypi/csvkit

heres an example one-liner that changes the source .txt from tsv
to csv (with `pyp`) and then to json (with csvkit's `csvjson`)
`$ cat bmi_girls_2_5_zscores.txt | pyp "p.replace('\t', ',')" | csvjson > bmifa_girls_2_5_zscores.json`
