Teaching and Learning Guide for: Climate‐Suicide Relationships: A Research Problem in Need of Geographic Methods and Cross‐Disciplinary Perspectives

Abstract

This guide accompanies the following article: Grady Dixon and Adam Kalkstein, ‘Climate‐Suicide Relationships: A Research Problem in Need of Geographic Methods and Cross‐Disciplinary Perspectives’, Geography Compass 3, 1961–1974, DOI: 10.1111/j.1749‐8198.2009.00286.xIntroductionSuicide is among the leading causes of mortality around the world, resulting in more deaths annually than homicide and war, combined. Interestingly, scientists have discovered a seasonal pattern in suicide incidence with higher numbers occurring in the late spring and early summer. While there is some evidence that climate might play a partial role in the observed seasonal cycle, the vast majority of climate‐suicide research has been conducted by mental health experts and others who have little background in geographic or climatological methods. Not surprisingly, the findings presented thus far on climate‐suicide relationships have often been misleading or confusing, highlighting the need for input from geographers and climatologists. Although mental health researchers are experts in their field of study, they are no more trained in geography than geographers are in mental health. Thus, this topic is one of many that illustrate the importance of collaborative efforts between experts from a variety of fields. Simply put, interdisciplinary research allows for more consistent, more accurate, and more relevant results when conducting scientific research. Other areas of climate‐health research have successfully blended efforts by scientists from both the medical and environmental realms, and much can be gained from following those examples.The following sections are meant to help educators develop a course that introduces climate‐suicide (and climate‐health) research to students. Suggested readings provide some history along with examples of previous research meant to elucidate common findings, problems, and goals of this topic. A list of relevant websites should help establish some of the priorities and motivating factors of the disciplines primarily involved in climate‐health research, and it might also be noted that there is an overall lack of attention given to climate‐suicide relationships in literature meant for the general public. Finally, 10 weeks of suggested course content and structure encourage the incorporation of larger science concepts into the course about climate and suicide. Example ‘focus questions’ follow the suggested course topics.Suggested Readings Durkheim, E., [1951] 1997: Suicide: A study in sociology. Simon and Schuster, pp. 104–122. This classic treatise by the famous French sociologist is still considered to be one of the most important works ever produced on the topic of suicide. Specifically, Chapter 3, entitled ‘Suicide and Cosmic Factors,’ addresses the plausibility, and ultimately the improbability, of climate significantly affecting suicide rates. Curtin, R. G., 1909: A study of the influence of climate upon suicide. Transactions of the American Clinical and Climatological Association, 25, 141–155. Available online at http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2262810 It is both educational and startling to survey the state of knowledge in this area 100 years ago. Despite the perfectly relevant title, readers should be prepared for some shocking conclusions based on racism, sexism, and environmental determinism. Deisenhammer, E. A., 2003: Weather and suicide: the present state of knowledge on the association of meteorological factors with suicidal behaviour. Acta Psychiatrica Scandinavica, 108, 402. DOI: 10.1046/j.0001‐690X.2003.00209.x A review of 27 published studies on attempted or completed suicide and weather and/or climate data. The papers discussed range in publication from 1973 to 2002 and include study periods ranging from 1 to 24 years (going back to 1954). Thirteen different countries are studied individually, and one project analyzed 62 different nations. Most apply statistical analyses as the primary method, and most relate suicide to temperature, precipitation, sunlight, and/or humidity. Partonen, T., J. Haukka, S. Pirkola, E. Isometsä, and J. Lönnqvist, 2004: Time patterns and seasonal mismatch in suicide. Acta Psychiatrica Scandinavica, 109, 110–115. DOI: 10.1046/j.0001‐690X.2003.00226.x The concept of ‘seasonal mismatch’ is employed to help predict periods of increased depression and suicide risk in Finland. This paper is important because it highlights the kind of complicating factors that suicide researchers must consider before analyzing their data, and the ambiguous results that should be expected even with the most rigorous methods. Lambert, G., C. Reid, D. Kaye, G. Jennings, and M. Esler, 2003: Increased suicide rate in the middle‐aged and its association with hours of sunlight. American Journal of Psychiatry, 160, 793–795 Noting the strong seasonality in suicide rates in southeastern Australia, this study finds the strongest relationship with suicide is displayed by direct sunlight rather than temperature or rainfall. This makes an interesting comparison for Nicholls et al. (2006), which claims that drought is related to suicide. Clearly, drought is likely to be accompanied by clear skies and more direct sunlight. Nicholls, N., C. D. Butler, and I. Hanigan, 2006: Inter‐annual rainfall variations and suicide in New South Wales, Australia, 1964–2001. International Journal of Biometeorology, 50, 139–143. This is one of very few published papers on this topic by a weather expert. Most studies are by psychologists, psychiatrists, or those in the health‐related fields. This paper is also interesting as a comparison to Lambert et al. (2003) as both focus on southeastern Australia, and their conclusions seem to be similar despite different messages. One has to wonder if either study is the victim of a false correlation due to multicollinearity. Dixon, P. G., A. N. McDonald, K. N. Scheitlin, J. E. Stapleton, J. S. Allen, W. M. Carter, M. R. Holley, D. D. Inman, and J. B. Roberts, 2007: Effects of temperature variation on suicide in five U.S. counties, 1991–2001. International Journal of Biometeorology, 51, 395–403. The results of this paper show no significant correlations between temperature and suicide in five locations with different latitudes, elevations, climate, and suicide patterns despite all locations showing some suicide seasonality. The paper also addresses the importance of using normalization techniques to separate seasonality and climate. Preti, A., G. Lentini, and M. Maugeri, 2007: Global warming possibly linked to an enhanced risk of suicide: Data from Italy, 1974–2003. Journal of Affective Disorders. The intriguing title of this paper is what makes it an educational read. Does the content of this paper justify the title? Have students discuss the implications of such titles.Suggested Websites 1. American Foundation for Suicide Prevention http://www.afsp.org/ This website is filled with information related to suicide prevention, statistics, and education. Most notably, the ‘About Suicide’ section provides facts and figures about suicide around the country and the world. The Frequently Asked Questions section gives some explanations about why certain groups of people are more prone to suicide than others. 2. World Health Organization http://www.who.int/globalchange/en/ While not specifically about suicide, this page addresses the effects of climate change on human health. This general topic is an important point of discussion to lead into, or stem from, analysis of climate‐suicide relationships. 3. Environmental Protection Agency http://www.epa.gov/climatechange/effects/health.html This site is also primarily about the effects of climate change on overall human health, but it addresses multiple components as a way to show how climate variation can yield significant changes in humans. 4. Intergovernmental Panel on Climate Change http://www.ipcc.ch/ The periodic reports compiled by this group of scientists and politicians are widely accepted as the state of understanding on the topic of climate change. The working group devoted to ‘Impacts, Adaptation, and Vulnerability’ is of particular interest for those studying climate effects on humans. 5. National Climatic Data Center http://www.ncdc.noaa.gov The National Climate Data Center (NCDC) is the official source of climate data for NOAA and it is the world’s largest archive of climate data. Educational institutions can view and/or download data and composites for stations all around the world.Sample Syllabus Week 1: Introduction to research on climate effects on human health Use WHO and EPA websites to highlight major issues. Have students discuss the ways they think climate is most likely to affect human health. Week 2: Early studies of environmental effects on humans Durkheim, E., [1951] 1997: Suicide: A study in sociology. Simon and Schuster, pp. 104–122.Curtin, R. G., 1909: A study of the influence of climate upon suicide. Transactions of the American Clinical and Climatological Association, 25, 141–155. Available online at http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2262810Point out some of the racist and sexist ideas that are common in very old research. Use these papers to explain the stigma of ‘environmental determinism,’ and why many future researchers chose to avoid environment‐health studies. Despite some culturally ignorant themes, do the papers provide utility and/or insight into the issues that we continue to study? Week 3: Introduction to past climate‐suicide research Deisenhammer, E. A., 2003: Weather and suicide: the present state of knowledge on the association of meteorological factors with suicidal behaviour. Acta Psychiatrica Scandinavica, 108, 402. DOI: 10.1046/j.0001‐690X.2003.00209.xHave students discuss possible reasons for differences in conclusions of so many scientific studies. Week 4: Importance of statistics in climate‐health studies Dixon, P. G., and A. J. Kalkstein, 2009: Climate‐suicide relationships: A research problem in need of geographic methods and cross‐disciplinary perspectives. Geography Compass, 4. DOI: 10.1111/j.1749‐8198.2009.00286.x.Sure, it would probably be better if scientists were analyzing specific physical and emotional variables as they respond to changes in weather. However, as alluded to in this paper, climate–suicide relationships are not understood well enough to be so specific. Statistical analyses are important for isolating potential key relationships that can be studied in further detail after sufficient evidence has been accumulated. It can sometimes be difficult just to identify consistent statistical relationships, and inconsistent methods further complicate the matter. Week 5: Separating the variables Dixon, P. G., A. N. McDonald, K. N. Scheitlin, J. E. Stapleton, J. S. Allen, W. M. Carter, M. R. Holley, D. D. Inman, and J. B. Roberts, 2007: Effects of temperature variation on suicide in five U.S. counties, 1991–2001. International Journal of Biometeorology, 51, 395–403.The utility of statistics can become trivial if efforts are not made to normalize the data and separate those that are likely to covary for reasons unrelated to suicide rates (i.e., temperature, humidity, and month). Week 6: Analysis of actual climate data Have students download and display a climate dataset from NCDC. Stress the importance of quality‐checking the data and illustrate how grouping/organization decisions can significantly affect the perception of the data being displayed in graphs, charts, etc. Weeks 7–8: The importance of multiple perspectives Lambert, G., C. Reid, D. Kaye, G. Jennings, and M. Esler, 2003: Increased suicide rate in the middle‐aged and its association with hours of sunlight. American Journal of Psychiatry, 160, 793–795Nicholls, N., C. D. Butler, and I. Hanigan, 2006: Inter‐annual rainfall variations and suicide in New South Wales, Australia, 1964–2001. International Journal of Biometeorology, 50, 139–143.These two studies both analyze climate‐suicide relationships in southeastern Australia over similar periods, yet they arrive at different conclusions. Further analysis reveals that their conclusions may be more related than they first appear. ‘Hours of sunlight’ and ‘rainfall variations’ certainly share a lot of variance. Perhaps these two sets of researchers should combine their efforts. Use this to show students the importance of using prior literature, multiple perspectives, and interdisciplinary collaborations. Week 9: Hidden and/or complicating factors Partonen, T., J. Haukka, S. Pirkola, E. Isometsä, and J. Lönnqvist, 2004: Time patterns and seasonal mismatch in suicide. Acta Psychiatrica Scandinavica, 109, 110–115. DOI: 10.1046/j.0001‐690X.2003.00226.xSuicide rates typically increase from winter through spring. Temperature also increases from winter through spring. Does this automatically mean that higher temperatures cause more suicides? This paper illustrates just one unexpected complicating factor that is much more likely to explain variations in suicide rates. It is these kinds of subtleties that researchers must try to find and explain. Week 10: Question everything Preti, A., G. Lentini, and M. Maugeri, 2007: Global warming possibly linked to an enhanced risk of suicide: Data from Italy, 1974–2003. Journal of Affective Disorders.To be good scientists, students should learn to question and analyze. Sometimes, they must do this while ignoring preconceived ideas. This week’s reading is a way for them to question a published scientific paper by well‐respected and widely published researchers. Also introduce some examples of ‘widely accepted’ concepts or myths that are easily contradicted as a way to show that the slightest bit of analysis can sometimes lead to important insight. The examples can be as simple as ‘lightning never strikes the same place twice,’ or as challenging as you want. The common ‘spring peak’ in suicide rates might be a good example since most people assume suicide rates increase during the ‘gloomy’ months of winter. Have the students think of similar examples.Focus Questions1. Why should we study climate‐suicide relationships?2. Is any single discipline better suited to study climate‐suicide relationships?3. What potential pitfalls should be avoided when comparing results of climate‐suicide studies from different countries? How can researchers successfully deal with these issues?4. What do you think is the ‘best’ resolution (spatial and temporal) for analyzing climate‐suicide relationships? Why?5. Think of two common variables that are clearly correlated, but not related by cause‐effect. How might you separate this covariance and test the strength of the causal relationship using statistics?