25 March 2024
Chris Johnston
[FYI request #25882 email]
Tēnā koe Chris
Your request for official information, reference: HNZ00039074 & HNZ00039162
Thank you for your emails on 26 February 2024, asking Health New Zealand | Te Whatu Ora for
the following under the Official Information Act 1982 (the Act):
HNZ00039162:
“1) Can you please confirm the total number of COVID vaccination records (doses) that
were covered by the dataset? Also break this down by Month and Dose Number.
So Col1 = Month of Dose, Col 2 = Dose Number, Cold 3 = Number Vaccinated. 10 people
given dose 3 in Feb 2023 would generate a record like "202302,3,10". The dataset would
have a final record that read: "Total, Total, <total number of doses>"
2) Please provide the number of people in the "Health Service User Population" as at the
end of each month from from 1 Jan 2019 to 31 Jan 2024 inclusive. Note that this should be
consistent with the definition requested in an earlier request in this OIA trail. This term was
used by Health NZ in its response so it needs to be quantified as a denominator.
3) Can you please confirm that the deaths in each category do not include the deaths in
previous dose categories. For example do the 169 deaths in 2022-11 for 41-60 years olds
after 3 doses that are marked "<=365" exclude the <5 deaths in the "<=180" category and
the <5 in the "<=90" category in the 4th column. The numbers in the Dose 4 category for
the same month seem to confirm that this is the case but we want to confirm (i.e. 8 for
"<365" is less than the the 24 Count for "<1=180".
4) There seems to be a discrepancy between the total deaths in this OIA and the stats at
this data source:
infoshare.stats.govt.nz
fyi.org.nz/request/25021.
This discrepancy grows in time as shown in a graph on these posts on X:
Can Health NZ please confirm that the total death stats are accurate compared to the
Infoshare site - especially if you are the source of the information at Stats as well as this
OIA?
5) Are the deaths up to 31 Dec 2023 inclusive rather than 1 Dec 2023 as stated in the
Summary tab of the OIA spreadsheet response?
AUGMENTED DATASET.
6) Can the following dataset based on Ticket Reference RT-268 please be provided.
Col1: Month of Death - from 1 Jan 2019 to 31 Jan 2024 inclusive
Col2: Age at Death - Redefine to "0 to 30", "31 to 60", "61 to 80", "81+".
Col3: Last COVID-19 dose number - keep as is
Col4: Days between last vaccination and death - leave as is
Col5: Count of Deaths - re-label
Col6: Total of Person-Days -
a) Where Dose = 1 to n: The number of "person-days" to the date of death for the people
who died in that month since their last dose. So if only 2 people died in a month for a Dose,
age group and timeframe then Col5 would be "<5" but if they dies 60 and 40 days after their
last dose then "person-days" would be 100.
b) Where Dose = 0: The number of person-days from 1 Jan 2019 to the date of death.
c) This does not reveal identity but allows %'s to be calculated. Age groups have also been
redefined/merged to minimise the occurrence of "<5"'s.
7) For the dataset in (6) please also provide additional rows for Age Group Totals. This is to
ensure that we can reconcile and validate the accuracy of the numbers being returned.
Therefore we expect rows like:
No Dose Summary: "Total,0 to 30, 0, Total, <Total Count of Deaths>,<Total Person-Days>"
At Least 1 Dose Summary: "Total,0 to 30, 1 to n, Total, <Total Count of Deaths>,<Total
Person-Days>"
... and similar for other Age Groups.
HNZ00039074:
Please provide a machine readable dataset in csv format under the same Principles and
Ways of Working as OIA.
A template dataset wil be the basis for providing results – one metric at a time. This should
enable a DBA/SQL analyst to write relatively efficient code.
TEMPLATE DATASET
Col1: Month of Dose - from 1 Jan 2019 to 31 Jan 2024 inclusive
Col2: Age at Dose - "0 to 30", "31 to 60", "61 to 80", "81+".
Col3: Last COVID-19 dose number: 0 where no dose was ever received.
Col4: Denominator:
a) If Col3=0 then Number of unvaccinated people in the “ealth Service User Population” as
at the end of the month in Col1
b) If Col3>0 then the Number of people vaccinated in the month in Col1
Col5: Events <=30 days – the metrix that occurs within 30 days from the last dose in Col 3
Col6: Events <=90 days – as per Col5 but for 31 to 90 days
Col7: Events <=180 days – as per Col5 but for 91 to 180 days
Col8: Events <=365 days – as per Col5 but for 181 to 365 days
Col9: Events 366+ days – as per Col5 but for more than 365 days
METRICS
Each of the following metrics should vbe delivered a seperate csv file.
M1) Miscarriages in Hospital (ICD-10 codes under O03)
a) Do not worry if there are numbers in 366+ as this may indicate a problem with a
subsequent pregnancy as a result of a dose.
M2) Person-Days of Mother from Last Dose
a) If Col3=0 then this is the number of person-days from 1 Jan 2019
b) If Col3>0 then this is the number of person-days from the last dose to the date of
admission
c) Do not worry if there are numbers in 366+ as this may indicate a problem with a
subsequent pregnancy as a result of a dose.
M3) Number of ER visitis (All Cause)
M4) ER visits (ICD-10 under I51 Cardiac)
M5) ER visits (ICD-10 under I60 through I69) Stroke Cranial
M6) Hospital Discharge Events – please specfiy whether this includes ER visits/discharges
M7) Hospital Inpatient Days
M8) Hospital Outpatient Events
M9) GP Visits
M10) Unique (Distinct) Count of People (NHI Numbers) across M1 to M9
NOTES:
N1) Suppression for <5 applies.
N2) Where a person generates a metric across multiple doses then the metrics that applied
to those dose periods need to be applied. So a hospital stay where a 2nd dose was
administered might a night beforehand (or even a previous admission) for for the first dose
and a few nights after the 2nd dose if there were further complications.
Response
As both requests were made by you on the same day and were asking for similar information, I
have combined the responses here.
On the 11th of March 2024, Health New Zealand reached out to you to attempt to refine these
requests, because we were unable to fulfil the requests as it would require substantial manual
collation of data.
On the 18th of March 2024, you responded without refining the requests as asked.
Therefore, I am refusing your requests under section 18(f) of the Act, as the information requested
cannot be made available without substantial collation and research. I have considered whether
fixing a charge for the supply of the information or extending the timeframe for response would
enable Te Whatu Ora to respond. I do not consider that either option would remove the impact that
supplying the information would have on our other operations.
How to get in touch
If you have any questions, you can contact us at
[email address].
If you are not happy with this response, you have the right to make a complaint to the
Ombudsman. Information about how to do this is available at
www.ombudsman.parliament.nz or
by phoning 0800 802 602.
As this information may be of interest to other members of the public, Health NZ may proactively
release a copy of this response on our website. Al requester data, including your name and
contact details, wil be removed prior to release.
Nāku iti noa, nā
Danielle Coe
Manager (OIA) Government Services
Health New Zealand | Te Whatu Ora
TeWhatuOra.govt.nz
Health NZ, PO Box 793,
Wel ington 6140, New Zealand
