Dear Ministry of Health,

Pursuant to the Official Information Act 1982, I would like to know:

1. How many individual's over the age of 18 were referred to publicly-funded head and neck (otolaryngology) services for tonsillectomy in each year of 2015-2024?
2. For the same time periods, how many referrals cited sleep apnea, diabetes, or asthma as extenuating factors?
3. For the same time periods, how many tonsillectomies were performed, grouped by extenuating factors (sleep apnea, diabetes, or asthma );

* Please breakdown the responses to questions 1-3 by health district (formerly district health boards) and age bands 18-19, 20-29, 30-39, 40-49, 50-59, 60-69, 70+.

Yours faithfully,

SAM C

Dear OIA Requests,

Can you please check that Health New Zealand accepted this request of transfer? I have received no update from them. I fear that if I am left to follow up with them, they will treat this as a new request - where they should have responded once you requested a transfer.

Yours sincerely,

SAMCNZ

Dear hnzOIA,

Thank you for your response. 

I accept that it may take some time to answer my questions. In saying that, I am unsatisfied with your response. Please see below for further details:

Note.
* International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) is used to classify diseases, injuries and related health problems (e.g., including comorbidities).
* Australian Classification of Health Interventions (ACHI) is used to classify surgeries, therapies and health interventions.

It appears that the “block code 412” [Tonsillectomy or adenoidectomy] referred to in your response is an ACHI term not a ICD10-AM term.

I am aware that individual hospitalisation events in the NDMS are coded with both ICD-10-AM diagnostic codes and ACHI procedure codes. You have chosen to only supply information relating to the ACHI procedure codes.

I am happy to limit the response to discharges of patients who received a “block code 412” [Tonsillectomy or adenoidectomy] procedure, but really do think that it is possible to provide details on the comorbidities I asked about.

REMEDY.
In the The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) First Edition to Eleventh Edition 2019 manual the following diagnostic codes are defined:

Diabetes (E10.x, type I diabetes mellitus; E11.x, type II diabetes mellitus).
Sleep apnea (G47.3x)
Asthma (J45.x)
(.x represents any trailing numbers in the specific codes)

I believe that it should be possible for you to update the previously supplied excel sheet of how many people with a “block 412” diagnostic code by year, age group, and DHB with stratification (i.e., those that either had, or did not have a diagnostic ICD-10-AM codes noted above for that hospitalisation event). I understand that these codes are not always available - but I am happy for you to provide the data with caveats, if you must.

Additionally, can you please provide clarification of whether these data relate to procedures performed at ‘public hospitals’ or those ‘funded by the public health system’. I am aware that some procedures ‘funded by the public health system’ are performed under contract at private hospitals.

Regards, 

SAMCNZ

P.S. the data provided is not confidential so can you please provide the file without password protection on the excel sheet. That would make it much easier for me to conduct the inter-regional analysis that I am ultimately after.

Kia ora Eden,

You do not seem to understand my request.

1. You have already compiled data relating to tonsillectomy/adenoidectomy procedures. This is great. (My comment was about the labelling of the data in the excel file, which refers to an ICD-10-AM block code 412 - which as you acknowledge, doesn't exist. It is in fact ACHI block code 412. We are in agreement).

2. I understand the structure of the NDMS dataset. It codes distinct hospitalisation events. You have identified the hospitalisation events that contain a procedure code relating to tonsillectomy/adenoidectomy procedures. This is a great first step of my request. What I am now asking for is that for those hospitalisation events that contains the relevant procedure codes - how many also contain ICD-10-AM codesfor:

Diabetes (E10.x, type I diabetes mellitus; E11.x, type II diabetes mellitus).
Sleep apnea (G47.3x)
Asthma (J45.x)
(.x represents any trailing numbers in the specific codes)

I do not understand how this is 'complex' and would incur a charge. I have done a similar query when I previously had access to NDMS data and it was not a complex request at all.

Yours sincerely,
SAMCNZ

Kia ora kotou,

Please excuse the terse tone of my earlier email. I had become frustrated this morning by the time it has taken to get an incomplete response -- after having initially made my request (to the Ministry of Health) for this data in January.

I have now taken some time to reflect on our full exchange and would like to apologise for my tone, unreservedly.

The point I was making about codes was that the first response that I received referred to an ICD-10-AM block code 412. I was unaware of the existence of ICD-10-AM block codes and this caused me a minor inconvenience as I do not have the ICD-10-AM/ACHI data dictionary and had to spend time deciphering this erroneous classification. Now, we both acknowledge that this should have been ACHI block code 412 (and now I acknowledge that it really wasn't too hard for me to confirm this).

As for the fullness of my request.

1. You have already compiled data relating to tonsillectomy/adenoidectomy procedures (ACHI block code 412). This is a great start - thank you.

2. As I recall, the NMDS records distinct hospitalisation events. I have previously observed this as a rectangular dataset - where each row represents a new hospitalisation event, and each column is a different variable, with values relevant to each row's hospitalisation event.

3. I presume that to create the initial dataset that you sent me, you filtered the NMDS for the presence of an ACHI procedure code from block code 412, grouped the results by year, age group and DHB, and then summarised/counted the rows.

4. What I am now asking for is, for those hospitalisation events (rows in the dataset) that contain the relevant procedure codes, how many also contain ICD-10-AM codes (in the diagnosis code columns. ) for:

* Diabetes (E10.x, type I diabetes mellitus; E11.x, type II diabetes mellitus).
* Sleep apnea (G47.3x)
* Asthma (J45.x)
(.x represents any trailing numbers in the specific codes)

(i.e., count those that have a procedure code relating to tonsillectomy/adenoidectomy procedures (ACHI block code 412) AND a diagnosis code relating to each of the comorbidities above).

I do not consider this request 'complex' - applying another filter to a rectangular dataset (and iterating over the different code sets) should still be 'simple'.

Note. This request is not for a formal research project - it is for personal interest. The commorbidities requested are those commonly cited by ENT specialists for why an adult may have a tonsillectomy/ adenoidectomy.

Yours sincerely,
SAMCNZ