PREgnancy-associated progression of chronic kidney DIsease: development of a Clinical predictive Tool (PREDICT):The National Registry of Rare Kidney Diseases (RaDaR), UK Renal Registry (UKRR), Hospital Episode Statistics(HES), and Maternity Services Data Set (MSDS) Linkage

Chronic kidney disease (CKD) is estimated to affect 3% of pregnant women in high-income countries, (Piccoli et al., 2018, #13860). It is estimated that 15 to 20 000 pregnancies per year in the UK are affected by CKD. However, this number is set to rise due to maternal factors associated with CKD (e.g., obesity and advancing maternal age) and accessibility of assisted conception. Current risk estimates for pregnancy-associated progression of moderate to severe reduction of renal function are based on a small single centre study (n=67). The study produced data from 1971 to 1997 regarding women with moderate to severe renal disease based on a classification that pre-dated the current classification of renal disease into CKD stages 1- 5. CKD 3 -5 covers a wider range of renal disease from mild to severe. In addition, for many women with rare renal diseases there are no data from sufficient numbers of women to inform likely pregnancy outcomes.

At the moment, there is no reliable way to work out which women with kidney disease are likely to have pregnancy-associated disease progression. The study team at King's College Hospital NHS Foundation Trust (KCL NHSFT) and the Department of Women and Children's Health at King's College London have established unique research collaborations to develop a prediction tool for pregnancy-associated progression of CKD using data from approximately 60,000 women. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London will develop a 'calculator' from information from thousands of pregnant women with kidney disease which can predict how much kidney function women are likely to lose in pregnancy. A preliminary predictive tool for pregnancy progression of CKD will be developed in pre-existing longitudinal cohorts of women recruited to the National Institute for Health Research (NIHR) portfolio studies (Pre-Eclampsia And Chronic Hypertension, rEnal and SLE (PEACHES) and Pregnancy Adaptation in Renal disease Study (PAIRS)).

AMENDMENT (v1.5)

To carry out this work, KCL NHSFT require additional HES APC data dating back from 1997/98 to latest available data. The need to get the extra years of data is in order to maximise the number of pregnancies that can be captured. The pregnancy rates in other cohorts studied has been lower than anticipated, resulting in the data period being extended. KCL NHSFT have explored numbers of pregnancies in women with CKD in two other population cohorts these included- Stockholm (2007-2017) and Kent (2010-2019) and the rate is approximately 33% of anticipated (<1% of pregnancies complicated by CKD compared with expected 3% of women affected). To ensure that KCL NHSFT are able to capture as many pregnancies as possible from the UKRR and RaDaR cohorts the study period has been extended from that which was requested under the pervious version of this agreement.

KCH NHSFT require HES APC data from as far back as 1997/98 for linkage with the UK Renal Registry and RaDaR cohort. The overall PREDICT study will be sufficiently powered, as KCH NHSFT have developed the model in separate cohorts, and anticipate there will be at least 6000 women in the RaDaR and UK RR dataset to enable validation (Ideally >3000 women needed). However, this wider data period will enable i) larger numbers of pregnancies in the cohorts of women with rare diseases to provide more meaningful descriptive data for these women ii) increase certainty of the maximum number of pregnancies to be identified. The MSDS data covers the period April 2015 - March 2019.

Of the 60,000 women provided in the cohort to NHS England, it is estimated that between approximately 750 and 6,000 cohort members will be identified as pregnant. Fertility rates* in women with CKD are very low. Through the study team's previous studies, they have identified only between 5 and 50 women per year with dialysis or kidney transplants to have pregnancies. Therefore, they have estimated that there will be very few captured and have estimated the final number to be between approximately 750 and 6,000 out of the original 60,000.

* Fertility rates as defined by Wiles K, Nelson-Piercy C, Bramham K Reproductive health and pregnancy in women with chronic kidney disease. Nat Rev Nephrol . 2018 Mar;14(3):165-184. doi: 10.1038/nrneph.2017.187.

In order, for predictive tools to be generalisable, multiple external validations in diverse patient groups including individual patient data from population cohorts are necessary. The calculator will then be refined and tested in this proposed data set (UK Rare Renal Diseases Registry (RaDaR)) by the UK RR team and the calculator's predictive performance will be shared with the PREDICT team.

Finally it will be externally validated with three international population data sets from UK (University of Kent Integrated Data), Canada (Ontario Renal Network) and Sweden (Stockholm CREAtinine Measurement (SCREAM). This work will be conducted in collaboration with international experts at the University of Toronto, University of Kent and Karolinska Institute in Stockholm. Collaborators will not have access or input to NHS England data or data processing. There will be no linkage between international data sets. These population datasets include comparable parameters to the RaDaR-UKRR dataset and will include pregnancy outcomes for women with kidney disease, laboratory data and baseline demographics. Kent data will be pseudonymised and accessed with a separate data sharing agreement with King's College London. SCREAM data and Ontario Renal Network data will be extracted and cleaned by the local teams, and the prediction tool shared for testing by local teams. This external validation exercise will not involve NHS England data.

King's College London's PREDICT team will share their prediction calculator with the Renal Association for testing on their dataset with serum creatinine linkage, and test performance will be shared. King's College London will not have access to any of the RaDaR-PREDICT linkage NHS England data and have no control over the methodology or data analysis. They are therefore not considered Data Processors or Data Controllers.

After the study team have developed the 'calculator' they will seek further funding to co-design an online platform or App with service users (Patients and Clinicians) which will be hosted by the UK Renal Registry.

The prediction calculator will allow women, their families and partners and health care professionals to input relevant data (e.g. severity of kidney disease, presence of high blood pressure) and the tool will estimate the chance of kidney function loss as result of pregnancy.

The study is funded by Kidney Research UK Stoneygate-Feehally grant which commenced in Oct 2018: 'PREgnancy-associated progression of chronic kidney disease: development of a Clinical predictive Tool (PREDICT)'

The UK Renal Registry (UKRR) is part of the Renal Association, a not-for-profit organisation registered with the Charity Commission (Registered in England No. 2229663). UKRR have high quality clinical databases open to requests from researchers through data collection from 71 adult and 13 paediatric renal centres. Participation is mandated in England through the NHS National Service Specification and the Chief Executive of each Trust is responsible for adherence to this contract. The majority of women receiving renal replacement therapy (haemodialysis, peritoneal dialysis and transplant) in the UK will be included in the UKRR database.

The National Registry of Rare Kidney Diseases (RaDaR) is a Renal Association initiative designed to pull together information from patients with certain rare kidney diseases and includes women with kidney disease who have had a pregnancy. Currently 101 NHS sites contributing to RaDaR, which includes all UK paediatric renal units and the majority of adult Renal centres. To date, over 25,000 patients have been recruited. The data are housed and managed using the same informatics infrastructure as the UK Renal Registry.

Linkage of NHS England data with UK Renal Registry data is required in order to generate a new data set for analysis to be used for validation of the prediction tool and to describe pregnancy outcomes for women with kidney disease. The study has approval from Research Ethics Committee and Confidential Advisory Group approval for data to be linked and used for this purpose.

GDPR LEGAL BASIS FOR DATA PROCESSING

Kings College Hospital NHS Foundation Trust’s GDPR Legal Basis for data processing is Article 6(1)(e) (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Kings College Hospital NHS Foundation Trust is a public authority. The proposal meets the Article 6 (1) (e) justifications for processing data due to the following public interest: Prediction of pregnancy outcomes was a top three research priority for women and their families who recently attended a RaDaR patient day. In a recent survey of women with chronic kidney disease (CKD), eight out of ten reported that accurate information about risk of kidney damage would influence their decision to conceive. In addition, 18 of 20 clinicians from the "Pregnancy and CKD Rare Renal Disease" group responded to a survey that a predictive tool would have considerable value in their practice. Kidney Care UK and Renal Association patient forums have confirmed that the study is in the public interest and would be welcomed by patients with CKD.

The Renal Association’s GDPR Legal Basis for data processing is Article 6(1)(f) (processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party except where such interests are overridden by the interests of fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child). The Renal Association shared a copy of its LIA with NHS England. NHS England have assessed this response against the ICO’s checklist (https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/lawful-basis-for-processing/legitimate-interests/) and are content that the requirements are met.

As Health data is a Special Category of Health data, both Data Controllers are additionally using Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject). The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in the below Section 5d.

The RA reliance on legitimate interest as a legal basis for the PREDICT study is justified by the potential benefits that the study will bring to different groups. Firstly, RA feel patients will benefit from the findings of this study, with the data being combined in this linkage providing a novel and unique insight into how kidney health impacts on the outcomes of pregnancy, allowing women to make informed choices about both their kidney and reproductive health when planning for a family. Secondly, the renal community will benefit from being able to better understand the relationships between kidney care and pre-/post-natal care to help them provide better outcomes for expectant mothers. Thirdly, both the RA and KCL benefit from the statistical insight and knowledge that working with this data will bring, improving the skill sets of their respective employees that will be benefit other areas of work, while also bringing in funding and exposure that might boost the future offerings of both the RA and KCL.

These benefits being as they are, all parties have an interest in the existing datasets being combined, allowing for analysis and insight with minimal impact on the privacy of patients, as the patients’ data have already been collected.

The PREDICT Study team will disseminate their findings to clinicians and researchers in the field by publishing the results in peer-reviewed journals and at international conferences. The study team hope to publish their findings in the following journals; Kidney International and Journal of American Society of Nephrology, and present their findings at the UK Kidney Week the American Society of Nephrology Kidney Week.

The PREDICT Study team will also disseminate findings to patients by launching the prediction tool on the RaDaR website, and invite patient charities including Kidney Research UK, Polycystic Kidney Disease Charity and Action on Pre-Eclampsia details to publish the results on their patient information pages. The PREDICT Study team will also publish findings on Twitter.

The PREDICT Study team aim that the research will be accessible to as broad an audience as possible, including researchers in the field as well as patients and clinicians and the media, with the hope that this will encourage further research and changes in clinical guidelines.

All published data will be aggregated, summarising all data into table and figures, with small number suppression applied as per the data set suppression rules for HES and MSDS.

The PREDICT Study team initially anticipated that preliminary model development results will be disseminated by the end of 2021, however, this is now planned for 2022. Subsequent development and validation was initially due to be published by the end of 2022, along with the prediction tool being made publicly available, however, this is now planned for end of 2023.

The PREDICT Study team will seek further funding to develop the tool interface for patients and clinicians. It is anticipated that the tool will be hosted on the RaDaR Pregnancy and CKD website. https://rarerenal.org/rare-disease-groups/pregnancy-and-chronic-kidney-disease-rdg/

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

The PREDICT Study team have nearly completed work on a risk prediction tool which is due to be published in early 2025.

Chronic kidney disease (CKD) is estimated to affect 3% of pregnant women in high-income countries, (Piccoli et al., 2018, #13860). It is estimated that 15 to 20 000 pregnancies per year in the UK are affected by CKD. However, this number is set to rise due to maternal factors associated with CKD (e.g., obesity and advancing maternal age) and accessibility of assisted conception. Current risk estimates for pregnancy-associated progression of moderate to severe reduction of renal function are based on a small single centre study (n=67). The study produced data from 1971 to 1997 regarding women with moderate to severe renal disease based on a classification that pre-dated the current classification of renal disease into CKD stages 1- 5. CKD 3 -5 covers a wider range of renal disease from mild to severe. In addition, for many women with rare renal diseases there are no data from sufficient numbers of women to inform likely pregnancy outcomes.

At the moment, there is no reliable way to work out which women with kidney disease are likely to have pregnancy-associated disease progression. The study team at King's College Hospital NHS Foundation Trust (KCL NHSFT) and the Department of Women and Children's Health at King's College London have established unique research collaborations to develop a prediction tool for pregnancy-associated progression of CKD using data from approximately 60,000 women. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London will develop a 'calculator' from information from thousands of pregnant women with kidney disease which can predict how much kidney function women are likely to lose in pregnancy. A preliminary predictive tool for pregnancy progression of CKD will be developed in pre-existing longitudinal cohorts of women recruited to the National Institute for Health Research (NIHR) portfolio studies (Pre-Eclampsia And Chronic Hypertension, rEnal and SLE (PEACHES) and Pregnancy Adaptation in Renal disease Study (PAIRS)).

AMENDMENT (v1.5)

To carry out this work, KCL NHSFT require additional HES APC data dating back from 1997/98 to latest available data. The need to get the extra years of data is in order to maximise the number of pregnancies that can be captured. The pregnancy rates in other cohorts studied has been lower than anticipated, resulting in the data period being extended. KCL NHSFT have explored numbers of pregnancies in women with CKD in two other population cohorts these included- Stockholm (2007-2017) and Kent (2010-2019) and the rate is approximately 33% of anticipated (<1% of pregnancies complicated by CKD compared with expected 3% of women affected). To ensure that KCL NHSFT are able to capture as many pregnancies as possible from the UKRR and RaDaR cohorts the study period has been extended from that which was requested under the pervious version of this agreement.

KCH NHSFT require HES APC data from as far back as 1997/98 for linkage with the UK Renal Registry and RaDaR cohort. The overall PREDICT study will be sufficiently powered, as KCH NHSFT have developed the model in separate cohorts, and anticipate there will be at least 6000 women in the RaDaR and UK RR dataset to enable validation (Ideally >3000 women needed). However, this wider data period will enable i) larger numbers of pregnancies in the cohorts of women with rare diseases to provide more meaningful descriptive data for these women ii) increase certainty of the maximum number of pregnancies to be identified. The MSDS data covers the period April 2015 - March 2019.

Of the 60,000 women provided in the cohort to NHS Digital, it is estimated that between approximately 750 and 6,000 cohort members will be identified as pregnant. Fertility rates* in women with CKD are very low. Through the study team's previous studies, they have identified only between 5 and 50 women per year with dialysis or kidney transplants to have pregnancies. Therefore, they have estimated that there will be very few captured and have estimated the final number to be between approximately 750 and 6,000 out of the original 60,000.

* Fertility rates as defined by Wiles K, Nelson-Piercy C, Bramham K Reproductive health and pregnancy in women with chronic kidney disease. Nat Rev Nephrol . 2018 Mar;14(3):165-184. doi: 10.1038/nrneph.2017.187.

In order, for predictive tools to be generalisable, multiple external validations in diverse patient groups including individual patient data from population cohorts are necessary. The calculator will then be refined and tested in this proposed data set (UK Rare Renal Diseases Registry (RaDaR)) by the UK RR team and the calculator's predictive performance will be shared with the PREDICT team.

Finally it will be externally validated with three international population data sets from UK (University of Kent Integrated Data), Canada (Ontario Renal Network) and Sweden (Stockholm CREAtinine Measurement (SCREAM). This work will be conducted in collaboration with international experts at the University of Toronto, University of Kent and Karolinska Institute in Stockholm. Collaborators will not have access or input to NHS Digital data or data processing. There will be no linkage between international data sets. These population datasets include comparable parameters to the RaDaR-UKRR dataset and will include pregnancy outcomes for women with kidney disease, laboratory data and baseline demographics. Kent data will be pseudonymised and accessed with a separate data sharing agreement with King's College London. SCREAM data and Ontario Renal Network data will be extracted and cleaned by the local teams, and the prediction tool shared for testing by local teams. This external validation exercise will not involve NHS Digital data.

King's College London's PREDICT team will share their prediction calculator with the Renal Association for testing on their dataset with serum creatinine linkage, and test performance will be shared. King's College London will not have access to any of the RaDaR-PREDICT linkage NHS Digital data and have no control over the methodology or data analysis. They are therefore not considered Data Processors or Data Controllers.

After the study team have developed the 'calculator' they will seek further funding to co-design an online platform or App with service users (Patients and Clinicians) which will be hosted by the UK Renal Registry.

The prediction calculator will allow women, their families and partners and health care professionals to input relevant data (e.g. severity of kidney disease, presence of high blood pressure) and the tool will estimate the chance of kidney function loss as result of pregnancy.

The study is funded by Kidney Research UK Stoneygate-Feehally grant which commenced in Oct 2018: 'PREgnancy-associated progression of chronic kidney disease: development of a Clinical predictive Tool (PREDICT)'

The UK Renal Registry (UKRR) is part of the Renal Association, a not-for-profit organisation registered with the Charity Commission (Registered in England No. 2229663). UKRR have high quality clinical databases open to requests from researchers through data collection from 71 adult and 13 paediatric renal centres. Participation is mandated in England through the NHS National Service Specification and the Chief Executive of each Trust is responsible for adherence to this contract. The majority of women receiving renal replacement therapy (haemodialysis, peritoneal dialysis and transplant) in the UK will be included in the UKRR database.

The National Registry of Rare Kidney Diseases (RaDaR) is a Renal Association initiative designed to pull together information from patients with certain rare kidney diseases and includes women with kidney disease who have had a pregnancy. Currently 101 NHS sites contributing to RaDaR, which includes all UK paediatric renal units and the majority of adult Renal centres. To date, over 25,000 patients have been recruited. The data are housed and managed using the same informatics infrastructure as the UK Renal Registry.

Linkage of NHS Digital data with UK Renal Registry data is required in order to generate a new data set for analysis to be used for validation of the prediction tool and to describe pregnancy outcomes for women with kidney disease. The study has approval from Research Ethics Committee and Confidential Advisory Group approval for data to be linked and used for this purpose.

GDPR LEGAL BASIS FOR DATA PROCESSING

Kings College Hospital NHS Foundation Trust’s GDPR Legal Basis for data processing is Article 6(1)(e) (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Kings College Hospital NHS Foundation Trust is a public authority. The proposal meets the Article 6 (1) (e) justifications for processing data due to the following public interest: Prediction of pregnancy outcomes was a top three research priority for women and their families who recently attended a RaDaR patient day. In a recent survey of women with chronic kidney disease (CKD), eight out of ten reported that accurate information about risk of kidney damage would influence their decision to conceive. In addition, 18 of 20 clinicians from the "Pregnancy and CKD Rare Renal Disease" group responded to a survey that a predictive tool would have considerable value in their practice. Kidney Care UK and Renal Association patient forums have confirmed that the study is in the public interest and would be welcomed by patients with CKD.

The Renal Association’s GDPR Legal Basis for data processing is Article 6(1)(f) (processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party except where such interests are overridden by the interests of fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child). The Renal Association shared a copy of its LIA with NHS Digital. NHS Digital have assessed this response against the ICO’s checklist (https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/lawful-basis-for-processing/legitimate-interests/) and are content that the requirements are met.

As Health data is a Special Category of Health data, both Data Controllers are additionally using Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject). The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in the below Section 5d.

The RA reliance on legitimate interest as a legal basis for the PREDICT study is justified by the potential benefits that the study will bring to different groups. Firstly, RA feel patients will benefit from the findings of this study, with the data being combined in this linkage providing a novel and unique insight into how kidney health impacts on the outcomes of pregnancy, allowing women to make informed choices about both their kidney and reproductive health when planning for a family. Secondly, the renal community will benefit from being able to better understand the relationships between kidney care and pre-/post-natal care to help them provide better outcomes for expectant mothers. Thirdly, both the RA and KCL benefit from the statistical insight and knowledge that working with this data will bring, improving the skill sets of their respective employees that will be benefit other areas of work, while also bringing in funding and exposure that might boost the future offerings of both the RA and KCL.

These benefits being as they are, all parties have an interest in the existing datasets being combined, allowing for analysis and insight with minimal impact on the privacy of patients, as the patients’ data have already been collected.

The PREDICT Study team will disseminate their findings to clinicians and researchers in the field by publishing the results in peer-reviewed journals and at international conferences. The study team hope to publish their findings in the following journals; Kidney International and Journal of American Society of Nephrology, and present their findings at the UK Kidney Week the American Society of Nephrology Kidney Week.

The PREDICT Study team will also disseminate findings to patients by launching the prediction tool on the RaDaR website, and invite patient charities including Kidney Research UK, Polycystic Kidney Disease Charity and Action on Pre-Eclampsia details to publish the results on their patient information pages. The PREDICT Study team will also publish findings on Twitter.

The PREDICT Study team aim that the research will be accessible to as broad an audience as possible, including researchers in the field as well as patients and clinicians and the media, with the hope that this will encourage further research and changes in clinical guidelines.

All published data will be aggregated, summarising all data into table and figures, with small number suppression applied as per the data set suppression rules for HES and MSDS.

The PREDICT Study team initially anticipated that preliminary model development results will be disseminated by the end of 2021, however, this is now planned for 2022. Subsequent development and validation was initially due to be published by the end of 2022, along with the prediction tool being made publicly available, however, this is now planned for end of 2023.

The PREDICT Study team will seek further funding to develop the tool interface for patients and clinicians. It is anticipated that the tool will be hosted on the RaDaR Pregnancy and CKD website. https://rarerenal.org/rare-disease-groups/pregnancy-and-chronic-kidney-disease-rdg/

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

No benefits to Health and Social Care have yet been yielded as no data has been disseminated under the previous version of this Agreement.

Chronic kidney disease (CKD) is estimated to affect 3% of pregnant women in high-income countries, (Piccoli et al., 2018, #13860). It is estimated that 15 to 20 000 pregnancies per year in the UK are affected by CKD. However, this number is set to rise due to maternal factors associated with CKD (e.g., obesity and advancing maternal age) and accessibility of assisted conception. Current risk estimates for pregnancy-associated progression of moderate to severe reduction of renal function are based on a small single centre study (n=67). The study produced data from 1971 to 1997 regarding women with moderate to severe renal disease based on a classification that pre-dated the current classification of renal disease into CKD stages 1- 5. CKD 3 -5 covers a wider range of renal disease from mild to severe. In addition, for many women with rare renal diseases there are no data from sufficient numbers of women to inform likely pregnancy outcomes.

At the moment, there is no reliable way to work out which women with kidney disease are likely to have pregnancy-associated disease progression. The study team at King's College Hospital NHS Foundation Trust (KCL NHSFT) and the Department of Women and Children's Health at King's College London have established unique research collaborations to develop a prediction tool for pregnancy-associated progression of CKD using data from approximately 60,000 women. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London will develop a 'calculator' from information from thousands of pregnant women with kidney disease which can predict how much kidney function women are likely to lose in pregnancy. A preliminary predictive tool for pregnancy progression of CKD will be developed in pre-existing longitudinal cohorts of women recruited to the National Institute for Health Research (NIHR) portfolio studies (Pre-Eclampsia And Chronic Hypertension, rEnal and SLE (PEACHES) and Pregnancy Adaptation in Renal disease Study (PAIRS)).

AMENDMENT (v1.5)

To carry out this work, KCL NHSFT require additional HES APC data dating back from 1997/98 to latest available data. The need to get the extra years of data is in order to maximise the number of pregnancies that can be captured. The pregnancy rates in other cohorts studied has been lower than anticipated, resulting in the data period being extended. KCL NHSFT have explored numbers of pregnancies in women with CKD in two other population cohorts these included- Stockholm (2007-2017) and Kent (2010-2019) and the rate is approximately 33% of anticipated (<1% of pregnancies complicated by CKD compared with expected 3% of women affected). To ensure that KCL NHSFT are able to capture as many pregnancies as possible from the UKRR and RaDaR cohorts the study period has been extended from that which was requested under the pervious version of this agreement.

KCH NHSFT require HES APC data from as far back as 1997/98 for linkage with the UK Renal Registry and RaDaR cohort. The overall PREDICT study will be sufficiently powered, as KCH NHSFT have developed the model in separate cohorts, and anticipate there will be at least 6000 women in the RaDaR and UK RR dataset to enable validation (Ideally >3000 women needed). However, this wider data period will enable i) larger numbers of pregnancies in the cohorts of women with rare diseases to provide more meaningful descriptive data for these women ii) increase certainty of the maximum number of pregnancies to be identified. The MSDS data covers the period April 2015 - March 2019.

Of the 60,000 women provided in the cohort to NHS Digital, it is estimated that between approximately 750 and 6,000 cohort members will be identified as pregnant. Fertility rates* in women with CKD are very low. Through the study team's previous studies, they have identified only between 5 and 50 women per year with dialysis or kidney transplants to have pregnancies. Therefore, they have estimated that there will be very few captured and have estimated the final number to be between approximately 750 and 6,000 out of the original 60,000.

* Fertility rates as defined by Wiles K, Nelson-Piercy C, Bramham K Reproductive health and pregnancy in women with chronic kidney disease. Nat Rev Nephrol . 2018 Mar;14(3):165-184. doi: 10.1038/nrneph.2017.187.

In order, for predictive tools to be generalisable, multiple external validations in diverse patient groups including individual patient data from population cohorts are necessary. The calculator will then be refined and tested in this proposed data set (UK Rare Renal Diseases Registry (RaDaR)) by the UK RR team and the calculator's predictive performance will be shared with the PREDICT team.

Finally it will be externally validated with three international population data sets from UK (University of Kent Integrated Data), Canada (Ontario Renal Network) and Sweden (Stockholm CREAtinine Measurement (SCREAM). This work will be conducted in collaboration with international experts at the University of Toronto, University of Kent and Karolinska Institute in Stockholm. Collaborators will not have access or input to NHS Digital data or data processing. There will be no linkage between international data sets. These population datasets include comparable parameters to the RaDaR-UKRR dataset and will include pregnancy outcomes for women with kidney disease, laboratory data and baseline demographics. Kent data will be pseudonymised and accessed with a separate data sharing agreement with King's College London. SCREAM data and Ontario Renal Network data will be extracted and cleaned by the local teams, and the prediction tool shared for testing by local teams. This external validation exercise will not involve NHS Digital data.

King's College London's PREDICT team will share their prediction calculator with the Renal Association for testing on their dataset with serum creatinine linkage, and test performance will be shared. King's College London will not have access to any of the RaDaR-PREDICT linkage NHS Digital data and have no control over the methodology or data analysis. They are therefore not considered Data Processors or Data Controllers.

After the study team have developed the 'calculator' they will seek further funding to co-design an online platform or App with service users (Patients and Clinicians) which will be hosted by the UK Renal Registry.

The prediction calculator will allow women, their families and partners and health care professionals to input relevant data (e.g. severity of kidney disease, presence of high blood pressure) and the tool will estimate the chance of kidney function loss as result of pregnancy.

The study is funded by Kidney Research UK Stoneygate-Feehally grant which commenced in Oct 2018: 'PREgnancy-associated progression of chronic kidney disease: development of a Clinical predictive Tool (PREDICT)'

The UK Renal Registry (UKRR) is part of the Renal Association, a not-for-profit organisation registered with the Charity Commission (Registered in England No. 2229663). UKRR have high quality clinical databases open to requests from researchers through data collection from 71 adult and 13 paediatric renal centres. Participation is mandated in England through the NHS National Service Specification and the Chief Executive of each Trust is responsible for adherence to this contract. The majority of women receiving renal replacement therapy (haemodialysis, peritoneal dialysis and transplant) in the UK will be included in the UKRR database.

The National Registry of Rare Kidney Diseases (RaDaR) is a Renal Association initiative designed to pull together information from patients with certain rare kidney diseases and includes women with kidney disease who have had a pregnancy. Currently 101 NHS sites contributing to RaDaR, which includes all UK paediatric renal units and the majority of adult Renal centres. To date, over 25,000 patients have been recruited. The data are housed and managed using the same informatics infrastructure as the UK Renal Registry.

Linkage of NHS Digital data with UK Renal Registry data is required in order to generate a new data set for analysis to be used for validation of the prediction tool and to describe pregnancy outcomes for women with kidney disease. The study has approval from Research Ethics Committee and Confidential Advisory Group approval for data to be linked and used for this purpose.

GDPR LEGAL BASIS FOR DATA PROCESSING

Kings College Hospital NHS Foundation Trust’s GDPR Legal Basis for data processing is Article 6(1)(e) (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Kings College Hospital NHS Foundation Trust is a public authority. The proposal meets the Article 6 (1) (e) justifications for processing data due to the following public interest: Prediction of pregnancy outcomes was a top three research priority for women and their families who recently attended a RaDaR patient day. In a recent survey of women with chronic kidney disease (CKD), eight out of ten reported that accurate information about risk of kidney damage would influence their decision to conceive. In addition, 18 of 20 clinicians from the "Pregnancy and CKD Rare Renal Disease" group responded to a survey that a predictive tool would have considerable value in their practice. Kidney Care UK and Renal Association patient forums have confirmed that the study is in the public interest and would be welcomed by patients with CKD.

The Renal Association’s GDPR Legal Basis for data processing is Article 6(1)(f) (processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party except where such interests are overridden by the interests of fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child). The Renal Association shared a copy of its LIA with NHS Digital. NHS Digital have assessed this response against the ICO’s checklist (https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/lawful-basis-for-processing/legitimate-interests/) and are content that the requirements are met.

As Health data is a Special Category of Health data, both Data Controllers are additionally using Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject). The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in the below Section 5d.

The RA reliance on legitimate interest as a legal basis for the PREDICT study is justified by the potential benefits that the study will bring to different groups. Firstly, RA feel patients will benefit from the findings of this study, with the data being combined in this linkage providing a novel and unique insight into how kidney health impacts on the outcomes of pregnancy, allowing women to make informed choices about both their kidney and reproductive health when planning for a family. Secondly, the renal community will benefit from being able to better understand the relationships between kidney care and pre-/post-natal care to help them provide better outcomes for expectant mothers. Thirdly, both the RA and KCL benefit from the statistical insight and knowledge that working with this data will bring, improving the skill sets of their respective employees that will be benefit other areas of work, while also bringing in funding and exposure that might boost the future offerings of both the RA and KCL.

These benefits being as they are, all parties have an interest in the existing datasets being combined, allowing for analysis and insight with minimal impact on the privacy of patients, as the patients’ data have already been collected.

The PREDICT Study team will disseminate their findings to clinicians and researchers in the field by publishing the results in peer-reviewed journals and at international conferences. The study team hope to publish their findings in the following journals; Kidney International and Journal of American Society of Nephrology, and present their findings at the UK Kidney Week the American Society of Nephrology Kidney Week.

The PREDICT Study team will also disseminate findings to patients by launching the prediction tool on the RaDaR website, and invite patient charities including Kidney Research UK, Polycystic Kidney Disease Charity and Action on Pre-Eclampsia details to publish the results on their patient information pages. The PREDICT Study team will also publish findings on Twitter.

The PREDICT Study team aim that the research will be accessible to as broad an audience as possible, including researchers in the field as well as patients and clinicians and the media, with the hope that this will encourage further research and changes in clinical guidelines.

All published data will be aggregated, summarising all data into table and figures, with small number suppression applied as per the data set suppression rules for HES and MSDS.

The PREDICT Study team initially anticipated that preliminary model development results will be disseminated by the end of 2021, however, this is now planned for 2022. Subsequent development and validation was initially due to be published by the end of 2022, along with the prediction tool being made publicly available, however, this is now planned for end of 2023.

The PREDICT Study team will seek further funding to develop the tool interface for patients and clinicians. It is anticipated that the tool will be hosted on the RaDaR Pregnancy and CKD website. https://rarerenal.org/rare-disease-groups/pregnancy-and-chronic-kidney-disease-rdg/

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

No benefits to Health and Social Care have yet been yielded as no data has been disseminated under the previous version of this Agreement.

Chronic kidney disease (CKD) is estimated to affect 3% of pregnant women in high-income countries, (Piccoli et al., 2018, #13860). It is estimated that 15 to 20 000 pregnancies per year in the UK are affected by CKD. However, this number is set to rise due to maternal factors associated with CKD (e.g., obesity and advancing maternal age) and accessibility of assisted conception. Current risk estimates for pregnancy-associated progression of moderate to severe reduction of renal function are based on a small single centre study (n=67). The study produced data from 1971 to 1997 regarding women with moderate to severe renal disease based on a classification that pre-dated the current classification of renal disease into CKD stages 1- 5. CKD 3 -5 covers a wider range of renal disease from mild to severe. In addition, for many women with rare renal diseases there are no data from sufficient numbers of women to inform likely pregnancy outcomes.

At the moment, there is no reliable way to work out which women with kidney disease are likely to have pregnancy-associated disease progression. The study team at King's College Hospital NHS Foundation Trust (KCL NHSFT) and the Department of Women and Children's Health at King's College London have established unique research collaborations to develop a prediction tool for pregnancy-associated progression of CKD using data from approximately 60,000 women. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London will develop a 'calculator' from information from thousands of pregnant women with kidney disease which can predict how much kidney function women are likely to lose in pregnancy. A preliminary predictive tool for pregnancy progression of CKD will be developed in pre-existing longitudinal cohorts of women recruited to the National Institute for Health Research (NIHR) portfolio studies (Pre-Eclampsia And Chronic Hypertension, rEnal and SLE (PEACHES) and Pregnancy Adaptation in Renal disease Study (PAIRS)).

AMENDMENT (v1.5)

To carry out this work, KCL NHSFT require additional HES APC data dating back from 1997/98 to latest available data. The need to get the extra years of data is in order to maximise the number of pregnancies that can be captured. The pregnancy rates in other cohorts studied has been lower than anticipated, resulting in the data period being extended. KCL NHSFT have explored numbers of pregnancies in women with CKD in two other population cohorts these included- Stockholm (2007-2017) and Kent (2010-2019) and the rate is approximately 33% of anticipated (<1% of pregnancies complicated by CKD compared with expected 3% of women affected). To ensure that KCL NHSFT are able to capture as many pregnancies as possible from the UKRR and RaDaR cohorts the study period has been extended from that which was requested under the pervious version of this agreement.

KCH NHSFT require HES APC data from as far back as 1997/98 for linkage with the UK Renal Registry and RaDaR cohort. The overall PREDICT study will be sufficiently powered, as KCH NHSFT have developed the model in separate cohorts, and anticipate there will be at least 6000 women in the RaDaR and UK RR dataset to enable validation (Ideally >3000 women needed). However, this wider data period will enable i) larger numbers of pregnancies in the cohorts of women with rare diseases to provide more meaningful descriptive data for these women ii) increase certainty of the maximum number of pregnancies to be identified. The MSDS data covers the period April 2015 - March 2019.

Of the 60,000 women provided in the cohort to NHS Digital, it is estimated that between approximately 750 and 6,000 cohort members will be identified as pregnant. Fertility rates* in women with CKD are very low. Through the study team's previous studies, they have identified only between 5 and 50 women per year with dialysis or kidney transplants to have pregnancies. Therefore, they have estimated that there will be very few captured and have estimated the final number to be between approximately 750 and 6,000 out of the original 60,000.

* Fertility rates as defined by Wiles K, Nelson-Piercy C, Bramham K Reproductive health and pregnancy in women with chronic kidney disease. Nat Rev Nephrol . 2018 Mar;14(3):165-184. doi: 10.1038/nrneph.2017.187.

In order, for predictive tools to be generalisable, multiple external validations in diverse patient groups including individual patient data from population cohorts are necessary. The calculator will then be refined and tested in this proposed data set (UK Rare Renal Diseases Registry (RaDaR)) by the UK RR team and the calculator's predictive performance will be shared with the PREDICT team.

Finally it will be externally validated with three international population data sets from UK (University of Kent Integrated Data), Canada (Ontario Renal Network) and Sweden (Stockholm CREAtinine Measurement (SCREAM). This work will be conducted in collaboration with international experts at the University of Toronto, University of Kent and Karolinska Institute in Stockholm. Collaborators will not have access or input to NHS Digital data or data processing. There will be no linkage between international data sets. These population datasets include comparable parameters to the RaDaR-UKRR dataset and will include pregnancy outcomes for women with kidney disease, laboratory data and baseline demographics. Kent data will be pseudonymised and accessed with a separate data sharing agreement with King's College London. SCREAM data and Ontario Renal Network data will be extracted and cleaned by the local teams, and the prediction tool shared for testing by local teams. This external validation exercise will not involve NHS Digital data.

King's College London's PREDICT team will share their prediction calculator with the Renal Association for testing on their dataset with serum creatinine linkage, and test performance will be shared. King's College London will not have access to any of the RaDaR-PREDICT linkage NHS Digital data and have no control over the methodology or data analysis. They are therefore not considered Data Processors or Data Controllers.

After the study team have developed the 'calculator' they will seek further funding to co-design an online platform or App with service users (Patients and Clinicians) which will be hosted by the UK Renal Registry.

The prediction calculator will allow women, their families and partners and health care professionals to input relevant data (e.g. severity of kidney disease, presence of high blood pressure) and the tool will estimate the chance of kidney function loss as result of pregnancy.

The study is funded by Kidney Research UK Stoneygate-Feehally grant which commenced in Oct 2018: 'PREgnancy-associated progression of chronic kidney disease: development of a Clinical predictive Tool (PREDICT)'

The UK Renal Registry (UKRR) is part of the Renal Association, a not-for-profit organisation registered with the Charity Commission (Registered in England No. 2229663). UKRR have high quality clinical databases open to requests from researchers through data collection from 71 adult and 13 paediatric renal centres. Participation is mandated in England through the NHS National Service Specification and the Chief Executive of each Trust is responsible for adherence to this contract. The majority of women receiving renal replacement therapy (haemodialysis, peritoneal dialysis and transplant) in the UK will be included in the UKRR database.

The National Registry of Rare Kidney Diseases (RaDaR) is a Renal Association initiative designed to pull together information from patients with certain rare kidney diseases and includes women with kidney disease who have had a pregnancy. Currently 101 NHS sites contributing to RaDaR, which includes all UK paediatric renal units and the majority of adult Renal centres. To date, over 25,000 patients have been recruited. The data are housed and managed using the same informatics infrastructure as the UK Renal Registry.

Linkage of NHS Digital data with UK Renal Registry data is required in order to generate a new data set for analysis to be used for validation of the prediction tool and to describe pregnancy outcomes for women with kidney disease. The study has approval from Research Ethics Committee and Confidential Advisory Group approval for data to be linked and used for this purpose.

GDPR LEGAL BASIS FOR DATA PROCESSING

Kings College Hospital NHS Foundation Trust’s GDPR Legal Basis for data processing is Article 6(1)(e) (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Kings College Hospital NHS Foundation Trust is a public authority. The proposal meets the Article 6 (1) (e) justifications for processing data due to the following public interest: Prediction of pregnancy outcomes was a top three research priority for women and their families who recently attended a RaDaR patient day. In a recent survey of women with chronic kidney disease (CKD), eight out of ten reported that accurate information about risk of kidney damage would influence their decision to conceive. In addition, 18 of 20 clinicians from the "Pregnancy and CKD Rare Renal Disease" group responded to a survey that a predictive tool would have considerable value in their practice. Kidney Care UK and Renal Association patient forums have confirmed that the study is in the public interest and would be welcomed by patients with CKD.

The Renal Association’s GDPR Legal Basis for data processing is Article 6(1)(f) (processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party except where such interests are overridden by the interests of fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child). The Renal Association shared a copy of its LIA with NHS Digital. NHS Digital have assessed this response against the ICO’s checklist (https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/lawful-basis-for-processing/legitimate-interests/) and are content that the requirements are met.

As Health data is a Special Category of Health data, both Data Controllers are additionally using Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject). The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in the below Section 5d.

The RA reliance on legitimate interest as a legal basis for the PREDICT study is justified by the potential benefits that the study will bring to different groups. Firstly, RA feel patients will benefit from the findings of this study, with the data being combined in this linkage providing a novel and unique insight into how kidney health impacts on the outcomes of pregnancy, allowing women to make informed choices about both their kidney and reproductive health when planning for a family. Secondly, the renal community will benefit from being able to better understand the relationships between kidney care and pre-/post-natal care to help them provide better outcomes for expectant mothers. Thirdly, both the RA and KCL benefit from the statistical insight and knowledge that working with this data will bring, improving the skill sets of their respective employees that will be benefit other areas of work, while also bringing in funding and exposure that might boost the future offerings of both the RA and KCL.

These benefits being as they are, all parties have an interest in the existing datasets being combined, allowing for analysis and insight with minimal impact on the privacy of patients, as the patients’ data have already been collected.

The PREDICT Study team will disseminate their findings to clinicians and researchers in the field by publishing the results in peer-reviewed journals and at international conferences. The study team hope to publish their findings in the following journals; Kidney International and Journal of American Society of Nephrology, and present their findings at the UK Kidney Week the American Society of Nephrology Kidney Week.

The PREDICT Study team will also disseminate findings to patients by launching the prediction tool on the RaDaR website, and invite patient charities including Kidney Research UK, Polycystic Kidney Disease Charity and Action on Pre-Eclampsia details to publish the results on their patient information pages. The PREDICT Study team will also publish findings on Twitter.

The PREDICT Study team aim that the research will be accessible to as broad an audience as possible, including researchers in the field as well as patients and clinicians and the media, with the hope that this will encourage further research and changes in clinical guidelines.

All published data will be aggregated, summarising all data into table and figures, with small number suppression applied as per the data set suppression rules for HES and MSDS.

The PREDICT Study team initially anticipated that preliminary model development results will be disseminated by the end of 2021, however, this is now planned for 2022. Subsequent development and validation was initially due to be published by the end of 2022, along with the prediction tool being made publicly available, however, this is now planned for end of 2023.

The PREDICT Study team will seek further funding to develop the tool interface for patients and clinicians. It is anticipated that the tool will be hosted on the RaDaR Pregnancy and CKD website. https://rarerenal.org/rare-disease-groups/pregnancy-and-chronic-kidney-disease-rdg/

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide

No benefits to Health and Social Care have yet been yielded as no data has been disseminated under the previous version of this Agreement.

Chronic kidney disease (CKD) is estimated to affect 3% of pregnant women in high-income countries, (Piccoli et al., 2018, #13860). It is estimated that 15 to 20 000 pregnancies per year in the UK are affected by CKD. However, this number is set to rise due to maternal factors associated with CKD (e.g. obesity and advancing maternal age) and accessibility of assisted conception. Current risk estimates for pregnancy-associated progression of moderate to severe reduction of renal function are based on a small single centre study (n=67). The study produced data from 1971 to 1997 regarding women with moderate to severe renal disease based on a classification that pre-dated the current classification of renal disease into CKD stages 1- 5. CKD 3 -5 covers a wider rage of renal disease from mild to severe. In addition, for many women with rare renal diseases there are no data from sufficient numbers of women to inform likely pregnancy outcomes.

At the moment, there is no reliable way to work out which women with kidney disease are likely to have pregnancy-associated disease progression. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London have established unique research collaborations to develop a prediction tool for pregnancy-associated progression of CKD using data from approximately 60,000 women. The study team at King's College Hospital NHS Foundation Trust and the Department of Women and Children's Health at King's College London will develop a 'calculator' from information from thousands of pregnant women with kidney disease which can predict how much kidney function women are likely to lose in pregnancy. A preliminary predictive tool for pregnancy progression of CKD will be developed in pre-existing longitudinal cohorts of women recruited to the National Institute for Health Research (NIHR) portfolio studies (Pre-Eclampsia And Chronic Hypertension, rEnal and SLE (PEACHES) and Pregnancy Adaptation in Renal disease Study (PAIRS)).

Of the 60,000 women provided in the cohort to NHS Digital, it is estimated that between approximately 750 and 6,000 cohort members will be identified as pregnant. Fertility rates* in women with CKD are very low. Through the study team's previous studies they have identified only between 5 and 50 women per year with dialysis or kidney transplants to have pregnancies. Therefore they have estimated that there will be very few captured, and have estimated the final number to be between approximately 750 and 6,000 out of the original 60,000.

* Fertility rates as defined by Wiles K, Nelson-Piercy C, Bramham K Reproductive health and pregnancy in women with chronic kidney disease. Nat Rev Nephrol . 2018 Mar;14(3):165-184. doi: 10.1038/nrneph.2017.187.

In order, for predictive tools to be generalisable, multiple external validations in diverse patient groups including individual patient data from population cohorts are necessary. The calculator will then be refined and tested in this proposed data set (UK Rare Renal Diseases Registry (RaDaR)) by the UK RR team and the calculator's predictive performance will be shared with the PREDICT team.

Finally it will be externally validated with three international population data sets from UK (University of Kent Integrated Data), Canada (Ontario Renal Network) and Sweden (Stockholm CREAtinine Measurement (SCREAM). This work will be conducted in collaboration with international experts at the University of Toronto, University of Kent and Karolinska Institute in Stockholm. Collaborators will not have access or input to NHS Digital data or data processing. There will be no linkage between international data sets. These population datasets include comparable parameters to the RaDaR-UKRR dataset and will include pregnancy outcomes for women with kidney disease, laboratory data and baseline demographics. Kent data will be pseudonymised and accessed with a separate data sharing agreement with King's College London. SCREAM data and Ontario Renal Network data will be extracted and cleaned by the local teams, and the prediction tool shared for testing by local teams. This external validation exercise will not involve NHS Digital data.

King's College London's PREDICT team will share their prediction calculator with the Renal Association for testing on their dataset with serum creatinine linkage, and test performance will be shared. King's College London will not have access to any of the RaDaR-PREDICT linkage NHS Digital data and have no control over the methodology or data analysis. They are therefore not considered Data Processors or Data Controllers.

After the study team have developed the 'calculator' they will seek further funding to co-design an online platform or App with service users (Patients and Clinicians) which will be hosted by the UK Renal Registry.

The prediction calculator will allow women, their families and partners and health care professionals to input relevant data (e.g. severity of kidney disease, presence of high blood pressure) and the tool will estimate the chance of kidney function loss as result of pregnancy.

The study is funded by Kidney Research UK Stoneygate-Feehally grant which commenced in Oct 2018: 'PREgnancy-associated progression of chronic kidney DIsease: development of a Clinical predictive Tool (PREDICT)'

The UK Renal Registry (UKRR) is part of the Renal Association, a not for profit organisation registered with the Charity Commission (Registered in England No. 2229663). UKRR have high quality clinical databases open to requests from researchers through data collection from 71 adult and 13 paediatric renal centres. Participation is mandated in England through the NHS National Service Specification and the Chief Executive of each Trust is responsible for adherence to this contract. The majority of women receiving renal replacement therapy (haemodialysis, peritoneal dialysis and transplant) in the UK will be included in the UKRR database.

The National Registry of Rare Kidney Diseases (RaDaR) is a Renal Association initiative designed to pull together information from patients with certain rare kidney diseases and includes women with kidney disease who have had a pregnancy. Currently 101 NHS sites contributing to RaDaR, which includes all UK paediatric renal units and the majority of adult renal centres. To date, over 25,000 patients have been recruited. The data are housed and managed using the same informatics infrastructure as the UK Renal Registry.

Linkage of NHS Digital data with UK Renal Registry data is required in order to generate a new data set for analysis to be used for validation of the prediction tool and to describe pregnancy outcomes for women with kidney disease. The study has approval from Research Ethics Committee and Confidential Advisory Group approval for data to be linked and used for this purpose.

GDPR LEGAL BASIS FOR DATA PROCESSING

Kings College Hospital NHS Foundation Trust’s GDPR Legal Basis for data processing is Article 6(1)(e) (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller). Kings College Hospital NHS Foundation Trust is a public authority. The proposal meets the Article 6 (1) (e) justifications for processing data due to the following public interest: Prediction of pregnancy outcomes was a top three research priority for women and their families who recently attended a RaDaR patient day. In a recent survey of women with Chronic Kidney Disease (CKD), eight out of ten reported that accurate information about risk of kidney damage would influence their decision to conceive. In addition, 18 of 20 clinicians from the "Pregnancy and CKD Rare Renal Disease" group responded to a survey that a predictive tool would have considerable value in their practice. Kidney Care UK and Renal Association patient forums have confirmed that the study is in the public interest and would be welcomed by patients with CKD.

The Renal Association’s GDPR Legal Basis for data processing is Article 6(1)(f) (processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party except where such interests are overridden by the interests of fundamental rights and freedoms of the data subject which require protection of personal data, in particular where the data subject is a child). The Renal Association shared a copy of its LIA with NHS Digital. We have assessed this response against the ICO’s checklist (https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/lawful-basis-for-processing/legitimate-interests/) and are content that the requirements are met.

As Health data is a Special Category of Health data, both Data Controllers are additionally using Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject). The data are required for research purposes in the public interest – meeting the conditions in the DPA 2018 Schedule 1 Part 1 (4) - which GDPR Recital 52(2) determines is an appropriate derogation from the prohibition on processing special categories of personal data. The ways in which the processing of data will be of benefit to the public – thereby demonstrating that the processing is in the public interest – are described in the below Section 5d.

The PREDICT Study team will disseminate their findings to clinicians and researchers in the field by publishing the results in peer-reviewed journals and at international conferences. The study team hope to publish their findings in the following journals; Kidney International and Journal of American Society of Nephrology, and present their findings at the UK Kidney Week the American Society of Nephrology Kidney Week.

The PREDICT Study team will also disseminate findings to patients by launching the prediction tool on the RaDaR website, and invite patient charities including Kidney Research UK, Polycystic Kidney Disease Charity and Action on Pre-Eclampsia details to publish the results on their patient information pages. The PREDICT Study team will also publish findings on Twitter.

The PREDICT Study team aim that the research will be accessible to as broad an audience as possible, including researchers in the field as well as patients and clinicians and the media, with the hope that this will encourage further research and changes in clinical guidelines.

All published data will be aggregated, summarising all data into table and figures, with small number suppression applied as per the data set suppression rules for HES and MSDS.

The PREDICT Study team anticipate that preliminary model development results will be disseminated by the end of 2020, and subsequent development and validation to be published by the end of 2021, along with the prediction tool made publicly available.

The PREDICT Study team will seek further funding to develop the tool interface for patients and clinicians. It is anticipated that the tool will be hosted on the RaDaR Pregnancy and CKD website. https://rarerenal.org/rare-disease-groups/pregnancy-and-chronic-kidney-disease-rdg/

Yielded Benefits is not a requirement for new applications.