dhis2-devs team mailing list archive

["Dapo Adejumo" <dapo_adejumo@xxxxxxxxx>]

Thanks Prosper for resuscitating this perennial issue.

The problem is real especially where you have a high turnover of health facilities ( mostly private health facilities) – and to complicated the problem, its is not only limited to “opening” and “closing” but also when health facilities move from one parent to another.

 

Calle, the idea of using a “null activity report” may work but the challenge is that most reporting rate calculations are based on complete dataset registrations which are removed from actual data value count. Using the option of “compulsory data elements” basis of completeness . However there are been many field challenges in defining these compulsory data elements.

Any great ideas to navigate this challenge will be great!

 

………………………………………

Regards,

Dapo Adejumo

+2348033683677

Skype : dapojorge

 

From: Dhis2-devs [mailto:dhis2-devs-bounces+dapo_adejumo=yahoo.com@xxxxxxxxxxxxxxxxxxx] On Behalf Of Calle Hedberg
Sent: 1 April, 2015 10:57 AM
To: Jason Pickering
Cc: Prosper BT; DHIS developers
Subject: Re: [Dhis2-devs] Added Orgunits/added datasets distorting old reporting rates

 

Hi,

 

The system should ONLY consider absolute (real) opening and closing dates, and should not accept any report for a reporting unit outside of this (what we generally call its "validity period").

 

reporting units with irregular activities must be handled "manually" - there are basically two options:

 

1. You exclude certain types of typically irregular operations from your completeness analysis. Typical cases would be irregular mobile units.

 

2. You capture a null activity report for irregular ops when they are inactive (e.g. enter a zero for the headcount data element or similar). That null report will thus be part of your completeness rate.

 

Regards

Calle

 

 

 

On 1 April 2015 at 11:34, Jason Pickering <jason.p.pickering@xxxxxxxxx <mailto:jason.p.pickering@xxxxxxxxx> > wrote:

Yeah, I do not know really how complicated it would be to implement in DHIS2, but I do know we have performed this type of analysis in other places outside of DHIS2 with SQL and scripts. Basically, it would involve getting all of the opening and closed dates of the facilities, and then the completed data set registrations and performing a period by period type of analysis on the reporting rates, to provide a dynamic denominator (i.e. number of facilities which are supposed to report). However, in practice, what we saw was that the opening/closed dates were often not really that accurate, often leading to percentages over 100%, since the facility was supposedly closed, but still submitted a report.

 

I think if it is that big of a problem (sounds bad), you should to consider to implement a custom report which would take this into account, but maybe the developers could comment on how difficult this would be to implement. 

 

Perhaps the bigger problem however is the focus on reporting compelteness, as opposed to what is actually in the data? :-/

 

Regards,

Jason

 

 

On Wed, Apr 1, 2015 at 11:25 AM, Prosper BT <ptb3000@xxxxxxxxx <mailto:ptb3000@xxxxxxxxx> > wrote:

Thanks Jason,

 

I know its complicated and very difficult to implement but in traditional national system focusing on reporting rates, it has come to point where districts would rather not add an Orgunit and if closed would want it deleted out of the systems or dataset removed.

 

May be Analytics should consider Opening data and closing date, so that new Orgunit are given a current opening date

 

Regards

 

 

 

On Wed, Apr 1, 2015 at 12:12 PM, Jason Pickering <jason.p.pickering@xxxxxxxxx <mailto:jason.p.pickering@xxxxxxxxx> > wrote:

This is really a quite intricate problem, if it is going to be done right. The easy "fix" would be to ignore whether a facility contrinbutes to completeness rates if the opening and closed dates are not valid. In practice however, it is more complicated than that, because a facility may only report a data set for a few months, or be open only parts of the year (I have seen this in two countries now, where certain facilities are always closed during certain months, i.e. the Himalayan winter and tropical rainy season). So, how would we account for this? The opening and closed dates really have no meaning in this case, because it is only open sometimes, and thus, is not expected to submit a repo

 

So, although it might fix the problem of adding new facilities and closing existing ones, I am not really so sure it would really solve the actual problem. 


Regards,

Jason

 

 

 

 

On Wed, Apr 1, 2015 at 11:02 AM, Calle Hedberg <calle.hedberg@xxxxxxxxx <mailto:calle.hedberg@xxxxxxxxx> > wrote:

Prosper,

 

HISP-SA is working on a few related issues:

- Validity periods (open/closeddates) for orgunits used to ensure no data entry for closed OUs, no validation rule violations for closed OUs, closed OUs not included in OrgUnitGroup counts used in indicators.

- Validity periods for indicators

- Validity periods for validation rules.

See relevant blueprints

 

I'm not sure if your issue is different - can you specify what you mean by "previous rates" in this context?

 

regards

calle

 

On 1 April 2015 at 10:46, Prosper BT <ptb3000@xxxxxxxxx <mailto:ptb3000@xxxxxxxxx> > wrote:

Hi Team,

 

Am not sure is am behind the decision of taking care of the challenge most districts are experiencing. 

 

What has been the decision so far.

 

Once a dataset is added to a facility, it affects the previous rates regardless of the opening data.

 

Whats is the status of this issues

 

Regards


 

-- 

Prosper Behumbiize, MPH
Phone:        +256 414 320076
Cell:             +256 772 139037 <tel:%2B256%20772%20139037> 

                     +256 752 751776 <tel:%2B256%20752%20751776> 
                    

 

_______________________________________________
Mailing list: https://launchpad.net/~dhis2-devs
Post to     : dhis2-devs@xxxxxxxxxxxxxxxxxxx <mailto:dhis2-devs@xxxxxxxxxxxxxxxxxxx> 
Unsubscribe : https://launchpad.net/~dhis2-devs
More help   : https://help.launchpad.net/ListHelp





 

-- 

*******************************************

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA 

Tel/fax (home): +27-21-685-6472 <tel:%2B27-21-685-6472> 

Cell: +27-82-853-5352 <tel:%2B27-82-853-5352> 

Iridium SatPhone: +8816-315-19274 <tel:%2B8816-315-19274> 

Email: calle.hedberg@xxxxxxxxx <mailto:calle.hedberg@xxxxxxxxx> 

Skype: calle_hedberg

*******************************************

 


_______________________________________________
Mailing list: https://launchpad.net/~dhis2-devs
Post to     : dhis2-devs@xxxxxxxxxxxxxxxxxxx <mailto:dhis2-devs@xxxxxxxxxxxxxxxxxxx> 
Unsubscribe : https://launchpad.net/~dhis2-devs
More help   : https://help.launchpad.net/ListHelp





 

-- 

Jason P. Pickering
email: jason.p.pickering@xxxxxxxxx <mailto:jason.p.pickering@xxxxxxxxx> 
tel:+46764147049 <tel:%2B46764147049> 





 

-- 

Prosper Behumbiize, MPH
Phone:        +256 414 320076
Cell:             +256 772 139037 <tel:%2B256%20772%20139037> 

                     +256 752 751776 <tel:%2B256%20752%20751776> 
                    





 

-- 

Jason P. Pickering
email: jason.p.pickering@xxxxxxxxx <mailto:jason.p.pickering@xxxxxxxxx> 
tel:+46764147049 <tel:%2B46764147049> 





 

-- 

*******************************************

Calle Hedberg

46D Alma Road, 7700 Rosebank, SOUTH AFRICA 

Tel/fax (home): +27-21-685-6472

Cell: +27-82-853-5352

Iridium SatPhone: +8816-315-19274

Email: calle.hedberg@xxxxxxxxx <mailto:calle.hedberg@xxxxxxxxx> 

Skype: calle_hedberg

*******************************************