1
|
Barak-Corren Y, Tsurel D, Keidar D, Gofer I, Shahaf D, Leventer-Roberts M, Barda N, Reis BY. The value of parental medical records for the prediction of diabetes and cardiovascular disease: a novel method for generating and incorporating family histories. J Am Med Inform Assoc 2023; 30:1915-1924. [PMID: 37535812 PMCID: PMC10654871 DOI: 10.1093/jamia/ocad154] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2023] [Revised: 07/16/2023] [Accepted: 07/24/2023] [Indexed: 08/05/2023] Open
Abstract
OBJECTIVE To determine whether data-driven family histories (DDFH) derived from linked EHRs of patients and their parents can improve prediction of patients' 10-year risk of diabetes and atherosclerotic cardiovascular disease (ASCVD). MATERIALS AND METHODS A retrospective cohort study using data from Israel's largest healthcare organization. A random sample of 200 000 subjects aged 40-60 years on the index date (January 1, 2010) was included. Subjects with insufficient history (<1 year) or insufficient follow-up (<10 years) were excluded. Two separate XGBoost models were developed-1 for diabetes and 1 for ASCVD-to predict the 10-year risk for each outcome based on data available prior to the index date of January 1, 2010. RESULTS Overall, the study included 110 734 subject-father-mother triplets. There were 22 153 cases of diabetes (20%) and 11 715 cases of ASCVD (10.6%). The addition of parental information significantly improved prediction of diabetes risk (P < .001), but not ASCVD risk. For both outcomes, maternal medical history was more predictive than paternal medical history. A binary variable summarizing parental disease state delivered similar predictive results to the full parental EHR. DISCUSSION The increasing availability of EHRs for multiple family generations makes DDFH possible and can assist in delivering more personalized and precise medicine to patients. Consent frameworks must be established to enable sharing of information across generations, and the results suggest that sharing the full records may not be necessary. CONCLUSION DDFH can address limitations of patient self-reported family history, and it improves clinical predictions for some conditions, but not for all, and particularly among younger adults.
Collapse
Affiliation(s)
- Yuval Barak-Corren
- Predictive Medicine Group, Computational Health Informatics Program, Boston Children’s Hospital, Boston, Massachusetts, USA
| | - David Tsurel
- Predictive Medicine Group, Computational Health Informatics Program, Boston Children’s Hospital, Boston, Massachusetts, USA
- Clalit Research Institute, Ramat Gan, Israel
- The Hebrew University of Jerusalem, Jerusalem, Israel
| | - Daphna Keidar
- Predictive Medicine Group, Computational Health Informatics Program, Boston Children’s Hospital, Boston, Massachusetts, USA
- Clalit Research Institute, Ramat Gan, Israel
| | - Ilan Gofer
- Clalit Research Institute, Ramat Gan, Israel
| | - Dafna Shahaf
- The Hebrew University of Jerusalem, Jerusalem, Israel
| | - Maya Leventer-Roberts
- Clalit Research Institute, Ramat Gan, Israel
- Department of Environmental Medicine and Public Health, Icahn School of Medicine at Mount Sinai, New York, New York, USA
- Department of Pediatrics, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Noam Barda
- Clalit Research Institute, Ramat Gan, Israel
| | - Ben Y Reis
- Predictive Medicine Group, Computational Health Informatics Program, Boston Children’s Hospital, Boston, Massachusetts, USA
- Harvard Medical School, Boston, Massachusetts, USA
| |
Collapse
|
2
|
Hämmerle M, Forer L, Schönherr S, Peters A, Grallert H, Kronenberg F, Gieger C, Lamina C. A Family and a Genome-Wide Polygenic Risk Score Are Independently Associated With Stroke in a Population-Based Study. Stroke 2022; 53:2331-2339. [PMID: 35387493 DOI: 10.1161/strokeaha.121.036551] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
BACKGROUND Positive family history and genetic risk scores have been shown to independently capture those individuals with high risk for stroke. The aim of our study was to evaluate the amount of shared information between family history and genetic risk and to investigate their combined effect on the association with prevalent and incident stroke cases. METHODS We obtained a family risk score (FamRS), weighted for disease onset and family size as well as genome-wide polygenic risk score (PGS) including over 3.2 million single-nucleotide polymorphisms in the population-based prospective KORA F3 (Cooperative Health Research in the Region of Augsburg) study (n=3071) from Southern Germany. FamRS and PGS were evaluated separately and combined. The measures were once treated as continuous variables but also divided in the highest 20%, 10%, 5%, and 1% percentiles. Odds ratios via logistic regression and hazard ratios via Cox regression were estimated. A stroke event was defined as a hospitalization for stroke that was self-reported in a standardized interview by certified and supervised personnel. RESULTS The FamRS outperformed other simplified family measures such as affected parents or number of affected family members. FamRS and PGS were not correlated, and no individuals were observed with both very high FamRS and very high PGS (top 1% percentile). In a combined model, both FamRS and PGS were independently from each other associated with risk of stroke, also independent of other traditional risk factors (p [FamRS]=0.02, p [PGS]=0.005). Individuals in the top 1% of either FamRS or PGS were found to have >5-fold risk for stroke (odds ratios, 5.82 [95% CI, 2.08-14]; P=0.0002). The results for incident stroke events showed the same trend but were not significant. CONCLUSIONS Our study shows that a family risk score and PGS capture different information concerning individual stroke risk. Combining the risk measures FamRS and PGS increases predictive power, as demonstrated in a population-based study.
Collapse
Affiliation(s)
- Michelle Hämmerle
- Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria (M.H., L.F., S.H., F.K., C.L.)
| | - Lukas Forer
- Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria (M.H., L.F., S.H., F.K., C.L.)
| | - Sebastian Schönherr
- Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria (M.H., L.F., S.H., F.K., C.L.)
| | - Annette Peters
- Institute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany (A.P., C.G., H.G.).,German Center for Diabetes Research (DZD), Neuherberg, Germany (A.P., C.G., H.G.).,German Research Center for Cardiovascular Research (DZHK), Partner Site Munich Heart Alliance, Munich, Germany (A.P.)
| | - Harald Grallert
- Institute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany (A.P., C.G., H.G.).,German Center for Diabetes Research (DZD), Neuherberg, Germany (A.P., C.G., H.G.)
| | - Florian Kronenberg
- Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria (M.H., L.F., S.H., F.K., C.L.)
| | - Christian Gieger
- Institute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany (A.P., C.G., H.G.).,German Center for Diabetes Research (DZD), Neuherberg, Germany (A.P., C.G., H.G.)
| | - Claudia Lamina
- Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria (M.H., L.F., S.H., F.K., C.L.)
| |
Collapse
|