Skip to main content

Publication, Part of

National Audit of Pulmonary Hypertension, 13th Annual Report

Audit, Open data

Publication Date:
19 Jan 2023
Geographic Coverage:
England, Scotland
Geographical Granularity:
NHS Trusts, Country, Hospital Trusts
Date Range:
01 Apr 2021 to 31 Mar 2022

Previous Chapter

National Standards

Current Chapter

Survival analysis

View all

Next Chapter

COVID-19-related mortality

Future publication format

Please note that future reports will be web-based only. This is to improve the accessibility of our publications.

19 January 2023 09:30 AM

Page contents

Survival analysis

Overview

Why is this important?
  • PH may be a life shortening condition. One of the aims of treatment is to prolong life as well as to improve its quality. Measurement of survival is used as a clinical outcome; it can be used to assess the effect of treatment on many life threatening diseases.
How is survival reported?
  • Kaplan Meier curves are used as a way of graphically displaying an estimate of the proportion of people who will still be in a defined group a number of days after a chosen start date. For example, it provides an estimate of the percentage of patients that will still be alive x years after they received a diagnosis (see below). 

Kaplan Meier curves graph displaying an estimate of the proportion of people who will still be in a defined group a number of days after a chosen start date

How is survival measured?
  • Survival is measured from the earliest diagnosis date for the patient’s latest diagnosis. 98% of patients keep the same PH diagnosis.
Who is included?
  • Only adult patients (1) within the cohort for longitudinal analysis are included (those whose first referral was on or after 1 April 2009 (2)). Patients whose diagnosis date is more than 1 year prior to first referral are excluded.
  • The survival curves are stopped if the number of patients used to calculate the cumulative survival drops below 50. This is to prevent the curve from artificially flattening due to patients whose mortality status cannot be confirmed.
Methodological change
  • The methodology underlying the survival curves has changed slightly since the 12th Annual Report. This has not affected the overall key findings. Further information can be found in the Methodological Change document accompanying this report (3).
How is mortality data collected?
  • Date of death is submitted directly to NAPH by the PH centres, then supplemented by mortality data from the Office for National Statistics (ONS).
  • Patients submitted by the Scottish PH centre (Golden Jubilee) cannot be traced using ONS data or National Records of Scotland data because they are pseudonymised in NAPH (2). Scotland are not included in the lung disease or left heart disease survival curves due to difficulties in obtaining death data for those patients.

Summary

Chart S 1 shows that patients with CTEPH have the best survival rates of all types of PH. Chart S 1 also shows that lung disease has the lowest survival rate with fewer than half likely to survive after 2 years following diagnosis.

Chart S 2 shows that survival for CTEPH patients that are operated is better than for those that are not operated on. Note that patients who have undergone surgery are younger at diagnosis than those who have not undergone surgery.

Survival analysis of patients with IPAH with comorbidities and IPAH without comorbidities is not included in the 13th Annual Report. This is because the NAPH team are in the process of incorporating the new NAPH comorbidities fields into the definition of IPAH with comorbidities and IPAH without comorbidities.

In previous annual reports IPAH with comorbidities is defined as a IPAH diagnosis with a secondary diagnosis relating to a comorbidity. The number of patients with IPAH with comorbidities recorded in the audit using this definition is believed to be an underestimate of the true prevalence and therefore survival rates for this cohort may not reflect the true rates of survival.

PH type – Survival curves

Chart S 1: Survival curves from diagnosis by PH type (n = 18,229),

Adult patients in the longitudinal cohort, Great Britain, 2009-22

Chart S 1: Survival curves from diagnosis by PH type (n = 18,229)

Notes:

1. For the number of patients at risk at each time point see the following section 'PH type – Annual survival'.

2. The PAH group includes patients with congenital heart disease APAH, connective tissue disease APAH, IPAH, portal hypertension APAH and other PAH sub-groups.

3. See section 'Overview' above for details on which patients are included and when the survival curve is stopped.

CTEPH – Survival curves

Chart S 2: Survival curves from diagnosis by CTEPH type (n = 3,780): comparison,

Adult patients in the longitudinal cohort, Great Britain, 2009-22 

Chart S 2: Survival curves from diagnosis by CTEPH type (n = 3,780): comparison

Notes:

1. For the number of patients at risk at each time point see the next section 'PH type - Annual survival'.

2. See section 'Overview' above for details on which patients are included and when the survival curve is stopped.

3. For a definition of confidence intervals see Glossary: Acronyms and definitions.

Key finding
  • Survival for CTEPH patients that are operated on is better than the survival for those that are not operated on.

PH type – Annual survival

Table S 1: Post-diagnosis survival by PH type,

Adult patients in the longitudinal cohort, Great Britain, 2009-22

Years  PAH Left heart disease Lung disease CTEPH Miscellaneous/ uncertain causes
Elapsed All Operated Not operated
  Patients % survival Patients % survival Patients % survival Patients % survival Patients % survival Patients % survival Patients % survival
0 6,170 100 4,542 100 3,033 100 3,780 100 1,640 100 2,140 100 704 100
1 4,899 86 3,464 84 1,715 65 3,114 90 1,527 96 1,587 86 491 77
2 3,951 74 2,745 72 1114 46 2,685 84 1,389 93 1,296 76 379 64
3 3,112 65 2,031 62 721 35 2,217 78 1,219 91 998 68 265 52
4 2,417 57 1,500 54 493 28 1,801 74 1036 88 765 61 204 44
5 1,885 51 1104 46 345 23 1,434 69 867 86 567 54 146 37
6 1,444 46 769 40 241 18 1161 65 726 83 435 48 111 31
7 1094 41 526 34 171 16 901 61 577 80 324 44 86 28
8 789 37 342 29 101 13 681 57 444 77 237 40 70 25
9 503 33 209 23 69 12 507 54 350 74 157 35 50 22
10 316 30 126 20 - - 350 51 251 71 99 33 - -
11 150 27 54 18 - - 197 47 139 68 58 27 - -
12 61 25 - - - - 78 44 60 65 - - - -

Notes:

- = Number of patients in the cohort below 50.

1. The PAH group includes patients with congenital heart disease APAH, connective tissue disease APAH, IPAH, portal hypertension APAH and other PAH sub-groups.

2. See the section 'Overview' for details on which patients are included and when the survival curve is stopped.

Key findings
  • Patients with CTEPH have the highest chance of survival of all types of PH, particularly those who have undergone an operation for the condition. Note that patients who have undergone surgery are younger at diagnosis than those who are not operated on.
  • Lung disease has the lowest survival rate with fewer than half likely to survive after 2 years following diagnosis.

PH type – Demographics

Table S 2: Demographics by PH type, Adult patients in the longitudinal cohort, Great Britain, 2009-22

Diagnostic group Patients Age at diagnosis Female Days after diagnosis on which the cohort had 50% mortality
Number Lower Median Upper Mean %
quartile quartile
PAH 6,170 46 61 71 58 68 5 years and 73 days
Left Heart Disease 4,542 67 74 79 72 64 4 years and 174 days
Lung Disease 3,033 60 68 75 67 50 1 years and 266 days
CTEPH
…All 3,780 52 66 75 63 50 10 years and 87 days
…Operated 1,640 48 60 70 58 45 Not reached within the study period
…Not operated 2,140 58 70 78 67 54 5 years and 194 days
Miscellaneous/uncertain causes 704 53 62 71 61 55 3 years and 91 days

Notes:

1. The PAH group includes patients with congenital heart disease APAH, connective tissue disease APAH, IPAH, portal hypertension APAH and other PAH sub-groups.

2. See the section 'Overview' above for details on which patients are included and when the survival curve is stopped.

Last edited: 19 January 2023 9:31 am

Previous Chapter

National Standards

Next Chapter

COVID-19-related mortality