Skip to main content
Key Specialties
Cardiology Diabetology Endocrinology Gastroenterology Geriatrics and Gerontology Gynecology and Obstetrics Hematology Infectious Disease Internal Medicine Nephrology Neurology Oncology Pediatrics Respiratory Medicine Rheumatology Surgery
Congress Coverage
2024 ACC Congress 2023 ESMO Congress 2023 EASD Congress 2023 ERS Congress 2023 ESC Congress 2023 EHA Congress 2023 ASCO Annual Meeting
Clinical Collections
Advances in Alzheimer’s

At a glance: The ONWARDS insulin icodec trials

A round-up of the ONWARDS phase 3 clinical trials evaluating the novel weekly insulin icodec in people with diabetes.
ADVANCED SEARCH
Log in
Top
BMC Medical Research Methodology
Published in: BMC Medical Research Methodology 1/2019

Open Access 01-12-2019 | Suicide | Research article

Explainable statistical learning in public health for policy development: the case of real-world suicide data

Authors: Paul van Schaik, Yonghong Peng, Adedokun Ojelabi, Jonathan Ling

Published in: BMC Medical Research Methodology | Issue 1/2019

Login to get access

Abstract

Background

In recent years, the availability of publicly available data related to public health has significantly increased. These data have substantial potential to develop public health policy; however, this requires meaningful and insightful analysis. Our aim is to demonstrate how data analysis techniques can be used to address the issues of data reduction, prediction and explanation using online available public health data, in order to provide a sound basis for informing public health policy.

Methods

Observational suicide prevention data were analysed from an existing online United Kingdom national public health database. Multi-collinearity analysis and principal-component analysis were used to reduce correlated data, followed by regression analyses for prediction and explanation of suicide.

Results

Multi-collinearity analysis was effective in reducing the indicator set of predictors by 30% and principal component analysis further reduced the set by 86%. Regression for prediction identified four significant indicator predictors of suicide behaviour (emergency hospital admissions for intentional self-harm, children leaving care, statutory homelessness and self-reported well-being/low happiness) and two main component predictors (relatedness dysfunction, and behavioural problems and mental illness). Regression for explanation identified significant moderation of a well-being predictor (low happiness) of suicide behaviour by a social factor (living alone), thereby supporting existing theory and providing insight beyond the results of regression for prediction. Two independent predictors capturing relatedness needs in social care service delivery were also identified.

Conclusions

We demonstrate the effectiveness of regression techniques in the analysis of online public health data. Regression analysis for prediction and explanation can both be appropriate for public health data analysis for a better understanding of public health outcomes. It is therefore essential to clarify the aim of the analysis (prediction accuracy or theory development) as a basis for choosing the most appropriate model. We apply these techniques to the analysis of suicide data; however, we argue that the analysis presented in this study should be applied to datasets across public health in order to improve the quality of health policy recommendations.
Appendix
Available only for authorised users
Footnotes
1
“a set of tools for modeling and understanding complex datasets” ([20], p. vii). We use the term ‘statistical learning’ rather than ‘machine learning’ as the former more accurately represents the statistical analysis used in this paper.
 
2
“rich source of indicators across a range of health and wellbeing themes that has been designed to support JSNA [Joint Strategic Needs Assessment] and commissioning to improve health and wellbeing, and reduce inequalities.” [34]
 
3
Excluded were the Isles of Scilly, City of London and Rutland because of limited data availability.
 
4
To allow for comparison of between the different approaches, the results of variance accounted for (R2) are presented for each approach.
 
5
In hierarchical and stepwise regression, the semi-partial correlation coefficient squared denotes additional variance explained by each predictor in the outcome variable, but in forced-entry regression it denotes unique variance explained.
 
6
Please note that the Fingertips repository does not contain data about interventions. Instead, here we analyse variables that could be targeted by interventions.
 
7
Potential covariates (such as the predictors from a stepwise regression model) that cannot be (directly) influenced by the intervention under consideration are not included in the model.
 
Literature
1.
Aísa R, Clemente J, Pueyo F. The influence of (public) health expenditure on longevity. Int J Public Health. 2014;59(5):867–75.PubMedCrossRef
2.
Bardsley M, Steventon A, Fothergill G: Untapped potential: Investing in health and care data analytics. 2019, 978–1–911615-30-9:.
3.
Barzilay S, Feldman D, Snir A, Apter A, Carli V, Hoven CW, Wasserman C, Sarchiapone M, Wasserman D. The interpersonal theory of suicide and adolescent suicidal behavior. J Affect Disord. 2015;183:68–74.PubMedCrossRef
4.
Bozeman SR, Hoaglin DC, Burton TM, Pashos CL, Ben-Joseph RH, Hollenbeak CS. Predicting waist circumference from body mass index. BMC Med Res Methodol. 2012;12(1):115.
5.
Breiman L. Statistical modeling: the two cultures. Stat Sci. 2001;16(3):199–215.CrossRef
6.
Choi SB, Lee W, Yoon J, Won J, Kim DW. Risk factors of suicide attempt among people with suicidal ideation in South Korea: A cross-sectional study. BMC Public Health. 2017;17(1):579.PubMedPubMedCentralCrossRef
7.
8.
Diez-Roux AV. Multilevel analysis in public health research. Annu Rev Public Health. 2000;21:171–92.PubMedCrossRef
9.
Dixon BE, Pina J, Kharrazi H, Gharghabi F, Richards J. What’s past is prologue: a scoping review of recent public health and global health informatics literature. Online J Public Health Inf. 2015;7(2):e216.
10.
11.
Field A. Discovering statistics using IBM SPSS statistics. 5th ed. London: Sage; 2017.
12.
Fox S, Flowers J. fingertipsR: Fingertips data for public health; 2018.
13.
Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, Musacchio KM, Jaroszewski AC, Chang BP, Nock MK. Risk factors for suicidal thoughts and behaviors: a meta-analysis of 50 years of research. Psychol Bull. 2017;143(2):187–232.PubMedCrossRef
14.
Gamache R, Kharrazi H, Weiner JP. Public and population health informatics: the bridging of big data to benefit communities. Yearb Med Inform. 2018;27(1):199–206.PubMedPubMedCentralCrossRef
15.
Ghani R, Foster I. Big data and social science: a practical guide to methods and tools. Boca Raton: CRC Press; 2017.
16.
17.
Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction, Springer series in statistics. New York: Springer; 2009.CrossRef
18.
Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. 2nd ed. USA: Guilford Press; 2017.
19.
Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3–12.PubMedCrossRef
20.
James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning with applications in R. New York: Springer; 2017.
21.
Kabacoff R. R in action. 2nd ed. Shelter Island: Manning; 2015.
22.
Kharrazi H, Lasser EC, AYasnoff W, Loonsk J, Advani A, Lehmann HP, Chin DC, Weiner JP. a proposed national research and development agenda for population health informatics: summary recommendations from a national expert workshop. J Am Med Inform Assoc. 2017;24(1):2–12.PubMedCrossRef
23.
24.
Kruschke J. Doing Bayesian data analysis. 2nd ed. London: Academic Press; 2015.
25.
Lee AS, Pan A, Harbarth S, Patroni A, Chalfine A, Daikos GL, Garilli S, Martínez JA, Cooper BS. Variable performance of models for predicting methicillin-resistant Staphylococcus aureus carriage in European surgical wards. BMC Infect Dis. 2015;15(1):105.PubMedPubMedCentralCrossRef
26.
MacKinnon DP. Introduction to statistical mediation analysis. New York: Erlbaum; 2008.
27.
Massoudi BL, Chester KG. Public health, population health, and epidemiology informatics: recent research and trends in the United States. Yearb Med Inform. 2017;26(1):241–7.PubMedPubMedCentralCrossRef
28.
Messer LC, Jagai JS, Rappazzo KM, Lobdell DT. Construction of an environmental quality index for public health research. Environ Health Global Access Sci Sour. 2014;13(1):39.
29.
Michie S, West R. Behaviour change theory and evidence: a presentation to government. Health Psychol Rev. 2013;7(1):1–22.CrossRef
30.
Murphy KR, Myors B. Testing the hypothesis that treatments have negligible effects: minimum-effect tests in the general linear model. J Appl Psychol. 1999;84(2):234–48.CrossRef
31.
Musci RJ, Kharrazi H, Wilson RF, Susukida R, Gharghabi F, Zhang A, Wissow L, Robinson KA, Wilcox HC. The study of effect moderation in youth suicide-prevention studies. Soc Psychiatry Psychiatr Epidemiol. 2018;53(12):1303–10.PubMedCrossRef
32.
Pedhazur E. Multiple regression in behavioral research: explanation and prediction. 3rd ed. London: Harcourt Brace; 1997.
33.
Pedhazur EJ, Schmelkin LP. Measurement, design and analysis: an integrated approach. Hillsdale; Hove: Lawrence Erlbaum; 1991.
34.
35.
Reps JM, Schuemie MJ, Suchard MA, Ryan PB, Rijnbeek PR. Design and implementation of a standardized framework to generate and evaluate patient-level prediction models using observational healthcare data. J Am Med Inform Assoc. 2018;25(8):969–75.PubMedPubMedCentralCrossRef
36.
Reynolds D, Hennessy E, Polek E. Is breastfeeding in infancy predictive of child mental well-being and protective against obesity at 9 years of age? Child Care Health Dev. 2014;40(6):882–90.PubMedCrossRef
37.
Rudin C: Please stop explaining black box models for high stakes decisions. 2018. arXiv preprint arXiv:1811.10154
38.
Samadder SR, Nagesh Kumar D, Holden NM. An empirical model to predict arsenic pollution affected life expectancy. Popul Environ. 2014;36(2):219–33.CrossRef
39.
Samaritans. Suicide statistics report 2017. Ewell, Surrey: Author; 2017.
40.
Sheldon KM. Integrating behavioral-motive and experiential-requirement perspectives on psychological needs: a two process model. Psychol Rev. 2011;118(4):552–69.PubMedCrossRef
41.
42.
Simon GE, Johnson E, Lawrence JM, Rossom RC, Ahmedani B, Lynch FL, Beck A, Waitzfelder B, Ziebell R, Penfold RB, Shortreed SM. Predicting suicide attempts and suicide deaths following outpatient visits using electronic health records. Am J Psychiatry. 2018;175(10):951–60.PubMedCrossRefPubMedCentral
43.
Tabachnick BG, Fidell LS. Using multivariate statistics. 6th ed. Boston, London: Pearson; 2013.
44.
Tan CL, Gan VBY, Saleem F, MAA H. Building intentions with the theory of planned behaviour: The mediating role of knowledge and expectations in implementing new pharmaceutical services in Malaysia. Pharm Pract. 2016;14(4):850.CrossRef
45.
Tu Y, Gunnell D, Gilthorpe MS. Simpson’s paradox, Lord’s paradox, and suppression effects are the same phenomenon - the reversal paradox. Emerg Themes Epidemiol. 2008;5:2.PubMedPubMedCentralCrossRef
46.
Veldkamp B. Mastering the data mass. Enschede: University of Twente; 2018.
47.
Wilcox H, Wissow L, Kharrazi H, Wilson R, Musci R, Zhang A, Robinson K. Data linkage strategies to advance youth suicide prevention. Evid Rep Technol Assess. 2016a;222(9):1–70.
48.
Wilcox HC, Kharrazi H, Wilson RF, Musci RJ, Susukida R, Gharghabi F, Zhang A, Wissow L, Robinson KA. Data linkage strategies to advance youth suicide prevention: a systematic review for a National Institutes of health pathways to prevention workshop. Ann Intern Med. 2016b;165(11):779–85.PubMedCrossRef
49.
Wilson NJ, Cordier R. A narrative review of Men's sheds literature: reducing social isolation and promoting men's health and well-being. Health Soc Care Community. 2013;21(5):451–63.PubMedCrossRef
Metadata
Title
Explainable statistical learning in public health for policy development: the case of real-world suicide data
Authors
Paul van Schaik
Yonghong Peng
Adedokun Ojelabi
Jonathan Ling
Publication date
01-12-2019
Publisher
BioMed Central
Published in
BMC Medical Research Methodology / Issue 1/2019
Electronic ISSN: 1471-2288
DOI
https://doi.org/10.1186/s12874-019-0796-7

Other articles of this Issue 1/2019

BMC Medical Research Methodology 1/2019 Go to the issue

Research article

Preventing bias from selective non-response in population-based survey studies: findings from a Monte Carlo simulation study

Research article

Illustration of different modelling assumptions for estimation of loss in expectation of life due to cancer

Research article

Sample size calculation for estimating key epidemiological parameters using serological data and mathematical modelling

Research Article

Risk-adjusted CUSUM control charts for shared frailty survival models with application to hip replacement outcomes: a study using the NJR dataset

Research article

How different online recruitment methods impact on recruitment rates for the web-based coortesnaweb project: a randomised trial

Research article

Why do participants drop-out: findings from a prospective pediatric cohort for fever surveillance established at Vellore, southern India