Skip Navigation LinksHome > May 2006 - Volume 17 - Issue 3 > Vulnerability to Heat-Related Mortality: A Multicity, Popula...
doi: 10.1097/01.ede.0000208477.36665.34
Original Article

Vulnerability to Heat-Related Mortality: A Multicity, Population-Based, Case-Crossover Analysis

Stafoggia, Massimo*; Forastiere, Francesco*; Agostini, Daniele†; Biggeri, Annibale‡; Bisanti, Luigi§; Cadum, Ennio∥; Caranci, Nicola¶; de'Donato, Francesca*; De Lisio, Sara†; De Maria, Moreno∥; Michelozzi, Paola*; Miglio, Rossella**; Pandolfi, Paolo†; Picciotto, Sally*; Rognoni, Magda§; Russo, Antonio§; Scarnato, Corrado†; Perucci, Carlo A.*

Free Access
Article Outline
Collapse Box

Author Information

From the *Department of Epidemiology, Rome E Health Authority, Rome, Italy; the †Epidemiological Observatory, Department of Public Health, Local Health Authority, Bologna, Italy; the ‡Department of Statistics, University of Florence, Florence, Italy; the §Epidemiological Observatory, Local Health Authority, Milan, Italy; ∥Epidemiological Services, Regional Environmental Protection Agency, Piedmont, Turin, Italy; ¶Epidemiological Services, Grugliasco 5 Health Authority, Turin, Italy; and the **Department of Statistical Sciences, University of Bologna, Bologna, Italy.

Submitted 26 August 2005; accepted 7 December 2005.

This work was partially supported by the Italian Department of Civil Protection (contract number: 491, 15/04/04).

Correspondence: Francesco Forastiere, Department of Epidemiology, Rome E Health Authority, Via Santa Costanza 53, 00198 Rome, Italy. E-mail: forastiere@asplazio.it.

Collapse Box


Background: Although studies have documented increased mortality during heat waves, little information is available on the subgroups most susceptible to these effects. We evaluated the effects of summertime high temperature on daily mortality among population subgroups defined by demographic characteristics, socioeconomic status, and episodes of hospitalization for various conditions during the preceding 2 years.

Methods: We studied a total of 205,019 residents of 4 Italian cities (Bologna, Milan, Rome, and Turin) age 35 or older who died during 1997–2003. The case-crossover design was applied to evaluate the association between mean apparent temperature (same and previous day) and all-cause mortality. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) of dying at 30°C (apparent temperature) relative to 20°C were estimated accounting for time, population changes, and air pollution.

Results: We found an overall OR of 1.34 (CI = 1.27–1.42) at 30°C relative to 20°C. The odds ratio increased with age and was higher among women (OR = 1.45; 1.37–1.52) and among widows and widowers (1.50; 1.33–1.69). Low area-based income modestly increased the effect. Among the preexisting medical conditions investigated, effect modification was detected for previous psychiatric disorders (1.69; 1.39–2.07), depression (1.72; 1.24–2.39), heart conduction disorders (1.77; 1.38–2.27), and circulatory disorders of the brain (1.47; 1.34–1.62). Temperature-related mortality was higher among people residing in nursing homes, and a large effect was also detected for hospitalized subjects.

Conclusions: Subsets of the population that are particularly vulnerable to high summer temperatures include the elderly, women, widows and widowers, those with selected medical conditions, and those staying in nursing homes and healthcare facilities.

Several studies have investigated the relationship between high temperature and mortality,1 both during specific heat waves2–4 and over a long time period, using modern time-series analysis5–8 or the case-crossover approach.9 A J-shaped relationship between daily temperature and all-cause mortality has been found, with an immediate time lag (same day or previous day at the most) of the heat effect.1,5,6,10 However, little information is available on the subgroups most vulnerable to the effects of hot temperatures, ie, those fractions of the population with a larger than average response either resulting from intrinsic susceptibility factors (such as clinical conditions) or higher exposure.11 Small-scale investigations2,3,12 suggest the following vulnerability factors: living alone among the elderly, having a low socioeconomic status, and being ill. However, comprehensive evaluation at the population level is lacking. In the United States, O'Neill and coworkers10 found that place of death (out of hospital), black race, and low educational level intensified the temperature–mortality relationship. Schwartz13 identified having been hospitalized for diabetes as an effect modifier for heat-related mortality. In Europe, the heat wave episode in summer 200312,14,15 has focused public health attention on heat-related mortality and the possible preventive actions to be introduced, especially among targeted population subgroups.

Our study was aimed at identifying specific conditions that render the individual particularly vulnerable to hot weather. We considered individual demographic characteristics, socioeconomic status, place of death, and chronic diseases of deceased subjects in 4 Italian cities. Extensive record linkage procedures were used to characterize subjects with respect to previous morbidity. We then applied the case-crossover approach, a method proposed to study triggers of acute events such as myocardial infarction16 and largely used in air pollution epidemiology17 to evaluate effect modification of the high temperature–mortality association.

Back to Top | Article Outline


Subjects and Individual Information

We considered subjects age 35 or older residing and dying in 4 Italian cities from all noninjury causes (International Classification of Diseases, 9th Revision [ICD-9]: 1–799) in the following periods: 2000–2003 in Bologna, 1999–2003 in Milan, 1998–2001 in Rome, and 1997–2003 in Turin. Death status was retrieved from Regional Registers of Causes of Death, which also include data on sex and age. Record linkage with city-specific population registers provided information on marital status (Milan and Turin only) and census block of residence (approximately 500 inhabitants per block). Median population income (family income in 1998) for each census block of residence, provided by the Ministry of Finance, was considered as an area-based indicator of socioeconomic status and divided into categories based on percentiles.

The city-specific mortality datasets were then linked (using individuals’ fiscal codes) with the regional hospital discharge files (which include hospitalizations in public and private hospitals nationwide of all resident citizens). All hospital admissions during the 2 years preceding death (excluding the last 28 days) were selected. We obtained information on both primary causes of admission and secondary contributing diagnoses, and classified each subject according to having been hospitalized for a list of 28 groups of diagnoses chosen a priori by adapting the Elixauser list of comorbidities.18 The 28-day exclusion was applied to distinguish between chronic conditions and a sudden deterioration of health in the few days before death. Information on hospitalization within 28 days of death was used only as part of the identification of the place of death. This variable was categorized as out-of-hospital (neither admission nor discharge in the last 4-week period), discharged 2–28 days before death, in-hospital, or in a nursing home (for Milan and Turin only).

Back to Top | Article Outline
Environmental Variables

Daily environmental data were obtained from the Italian Air Force Meteorological Service, which provided temperature, relative humidity/dew point temperature, and barometric pressure measured at the nearest city airport. We used daily mean apparent temperature as the exposure variable.19 This combination of air temperature and dew point temperature represents physical stress deriving from extreme summer conditions better than does air temperature alone. The average exposure on the day of death and on the day before (lag 0–1) was used, because many studies found a short latency of the effect of high temperatures on mortality.1,5,6,10 We also collected daily data on particulate matter with aerodynamic diameter lower than 10 μm (PM10; lag 0–1) and ozone (daily maximum 8-hour running mean during May–September) from the Regional Environmental Protection Agencies, because air pollution has been associated with short-term increase in mortality.20–22 Urban background city monitors provided these latter variables.

Back to Top | Article Outline
Data Analysis

Statistical analysis was performed in 3 stages. First, the concentration–response curve of the relationship between apparent temperature and noninjury mortality was explored for each city using the case-crossover design.16 Control periods were selected using the “time-stratified” approach,23 in which the study period was divided into monthly strata, and control days for each case were selected as the same days of the week in the stratum. A conditional logistic regression analysis was performed for each city, modeling the exposure variable as a cubic penalized spline. The numbers of knots, and the smoothness of the curves, were chosen to minimize the Akaike information criterion (AIC) index. The final smoothness was tuned to avoid overfitting. For each location, the regression model controlled for the confounding effects of temporary population decrease in the summer period, holidays, influenza epidemics, linear terms for PM10 (lag 0–1), and barometric pressure (lag 0). Long-term and seasonal time trends, as well as day of the week, were controlled for by design. The role of summer ozone as a potential confounder was evaluated in a sensitivity analysis. All analyses were performed with R software version 2.1.0 (R Foundation for Statistical Computing, http://www.R-project.org).

We inspected the 4 plots of the temperature–mortality association to identify 2 city-specific cut points of the J curve: the level of apparent temperature when mortality starts to increase in a nonlinear way and the point at which the temperature–mortality relationship assumes a steep linear trend. The objective was to approximate the smoothed curve into 3 linear splines to simplify the overall relationship. Alternative models were inspected with a higher number of degrees of freedom and different location of the knots, but the model with 3 linear splines turned out to be the best in terms of AIC index.

In the second stage, the analysis of effect modification was performed in each location, approximating the apparent temperature–mortality relationship with 3 linear splines with 2 inner knots according to the city-specific cut points. The effect of apparent temperature at 30°C versus 20°C was then estimated.

Third, city-specific results were combined in a meta-analysis and potential heterogeneity was explored using random-effects models. The maximum likelihood method was used.24 All results are expressed as pooled odds ratios (ORs), with 95% confidence intervals (CIs), of dying on a day with 30°C apparent temperature relative to 20°C. Effect modification was tested and results are reported as the relative effect modification (REM) index calculated as the ratio between the specific odds ratio and the odds ratio from the reference category.

Back to Top | Article Outline


Table 1 displays a summary of the environmental variables considered and the number of deaths included in the analysis for each city. Overall, mean apparent temperature was highest in Rome and lowest in Turin. Milan had the greatest variability and the most extreme values in apparent temperature, whereas Rome's distribution was less dispersed. Table 1 also reports the distribution of the difference between apparent temperature in cases and controls in that the case-crossover design is focused on the within-subject variation and so the relevant exposure variable relates to the variability among case periods and control periods rather than the exposure of the cases alone.25 The 4 cities seem to be quite similar, with Rome showing the smallest variability. A total of 205,019 deaths were included in the study with the largest contribution from Rome.

Table 1
Table 1
Image Tools

Figure 1 shows the concentration–response curves between daily mean apparent temperature (lag 0–1) and noninjury mortality for the 4 locations. All 4 curves are J-shaped, with slightly different turning points (20°C and 26°C for Bologna, 22°C and 29°C for Milan, 20°C and 26°C for Rome, and 23°C and 27°C for Turin) and different slopes in the right arm (steeper in Milan and Turin than in Bologna and Rome). The effect of apparent temperature was approximately zero in the left arm of the concentration–response curves, but apparent temperature was modeled at immediate lag, whereas the cold effect has usually a much higher latency, up to several weeks.

Figure 1
Figure 1
Image Tools

The city-specific results, expressed as odds ratio of dying on days with 30°C in mean apparent temperature (lag 0–1) versus days with 20°C were: Bologna 1.37 (95% CI = 1.22–1.54), Milan 1.27 (1.19–1.53), Rome 1.30 (1.22–1.39), and Turin 1.45 (1.37–1.54). When the city-specific results were combined, we found an overall OR of 1.34 (1.27–1.42). In Table 2, combined effect estimates by age and sex are presented from either fixed or random-effects models. The odds ratios for women were higher than for men in each age category. In addition, there was a clear increasing trend of harmful effect of high temperature on mortality with age for both men and women.

Table 2
Table 2
Image Tools

Table 3 shows the combined results for the overall population and by sex, age, marital status, area-based income, previous hospital admissions, and place of death. Relative effect modification indexes and exact P values are also reported. A greater OR was found among widowed, unmarried, and divorced subjects (OR = 1.50; 95% CI = 1.33–1.69) than among married subjects (1.21; 1.13–1.28). No effect modification for area-based income was detected, although the OR was slightly lower among those in the highest quintile of the distribution. Subjects who had been hospitalized in the 2 preceding years had a smaller effect than those who had not. Place of death was an important effect modifier, because those discharged from a hospital 2–28 days before death had a reduced heat-related mortality, whereas people who were in a nursing home were more susceptible. In-hospital and out-of-hospital deaths had a similar association with mean apparent temperature, and there was an increase in mortality for both long-term care patients (more than 59 days) and those whose hospital stay lasted less than 60 days.

Table 3
Table 3
Image Tools

Table 4 shows the combined results for 28 groups of diagnoses as a primary or secondary cause of hospital admission in the 2 years before death. Psychoses, depression, and conduction disorders of the heart were effect modifiers, with ORs of 1.70 (CI = 1.39–2.09), 1.71 (1.23–2.38), and 1.77 (1.38–2.27), respectively. Those with previous cerebrovascular diseases (12% of the total cases) also had a higher risk (1.46; 1.33–1.61) than those not affected. On the other hand, subjects with cancer had a lower relative risk of dying during hot days than subjects without cancer (1.20; 1.13–1.28). No noticeable positive or negative effect modification was detected for other conditions. The main findings of the study are summarized in Figure 2.

Table 4
Table 4
Image Tools
Figure 2
Figure 2
Image Tools

A number of sensitivity analyses have been conducted and the main results are reported in Table 5. Age could be a confounder responsible for some of the apparent effect modification. However, it was difficult in our model to adjust for age because 3 age–temperature interaction terms would be needed in addition to the main effect modifier under focus, and the interpretation of the coefficients would not be straightforward. Therefore, we repeated all the analyses with a restriction to the population age 65+ years or age 75+ years. The results were quite similar to those found in the 35+ years age category, although the power was reduced especially in the 75+ years age group. In addition, the results were robust when the 2003 data were excluded from the analysis, indicating that the effects found were not merely the result of a specific heat wave. Finally, the inclusion of summer ozone as a confounder in the city-specific models did not change the risk estimates in a meaningful way.

Table 5
Table 5
Image Tools
Back to Top | Article Outline


We designed the study to identify subgroups vulnerable to heat, addressing a specific request from public health authorities seeking to better target social and medical intervention. We found that the following categories of people were at higher risk of dying on hot days: elderly, women, widows and widowers, and people with psychiatric conditions, depression, heart conduction disorders, and previous cerebrovascular diseases. People who were in nursing homes or in the hospital, and thus receiving greater social or medical attention, did not seem to be protected against heat-related mortality.

The adverse effect of high temperatures on mortality has been well documented in the United States,6,9,26 Europe,7,27 and Italy.14 We found an overall 34% increase of risk on days with mean apparent temperature of 30°C versus days with 20°C, similar to results in the other European studies but higher than U.S. estimates. O'Neill et al10 found an overall 5% increase of risk when temperatures rise from 15°C to 29°C, much less than in the present study. However, most of the excess in their study was in out-of-hospital deaths (+10%), whereas no excess was observed for in-hospital mortality. The greater availability of air conditioning in the U.S. hospitals may reduce the excess of risk attributable to heat.28

Several studies found a higher effect of hot temperatures on mortality in the elderly,26,27,29,30 and our results confirm the increasing trend of risk with age. Decreased sweating31 and difficulty in thermoregulation with age32 are the most important physiopathological factors. Our results, however, show an increase of risk even for the younger age group (35–64 years old), especially among women. O'Neill et al10 found an excess mortality (at 29°C) for the <65 age group but no differences between the sexes. Other studies reported sex differences, with a higher excess in men during the 1995 heat wave in Chicago33 and a higher excess in women in the 1995 heat wave in London.4 In other cases, no effect modification by sex was found.34,35 Vassallo et al36 conducted a study among elderly patients living in an institution and found a higher risk of marginal hyperthermia in women than in men.

Excess mortality resulting from high temperatures has already been found in people residing in nursing homes with no air conditioning.37–39 We found a greater risk for subjects living in nursing homes (OR = 1.61; CI = 1.41–1.84), which are mostly without air conditioning in Italy. In addition, we found that people who were in the hospital were also at risk for dying because of high temperatures (1.32; 1.27–1.40). A larger number of critically ill patients hospitalized during the heat wave might be the easiest interpretation of the finding. Our analysis, however, shows an increased heat-related mortality even among those who were in the hospital before the heat wave (complete data not shown, available on request). Thus, air conditioners in health facilities may provide a means for the prevention of heat-related mortality.

The analysis of hospital admissions in the 2 years before death suggests that persons with specific chronic conditions are especially vulnerable to hot temperatures. Mortality was higher in patients affected by depression and psychiatric conditions, perhaps because of the use of medicines that alter thermoregulation ability.38 The increased risk of dying of cerebrovascular diseases as a consequence of extreme heat has already been reported,40 whereas our result of an augmented risk for patients affected by conduction disorders seems to be a new finding. Patients might need an increased heart rate during hot days, and if unable to support this need, this may lead to a fatal cardiac crisis.

The present study also found a risk of dying for people who were not hospitalized in the 2 years before death. The odds ratio in this group was 1.42 (CI = 1.34–1.51), whereas in hospitalized subjects, it was 1.31 (1.23–1.39). Thus, having been admitted to the hospital in the 2-year period before death (excluding last 28 days) does not seem to be a marker of susceptibility except for specific pathologies. Because the case-crossover design is only able to estimate relative effects, this result should not be interpreted as a protective effect but as a less-than-multiplicative one. In fact, the absolute risk of mortality for previously hospitalized subjects is presumably much higher than the risk for the nonhospitalized, and thus the relative contribution of apparent temperature on mortality turns out to be smaller in the first group.

Several strengths of the present study deserve consideration. This study involved 4 cities and more than 200,000 deaths in a fairly recent period that includes, for 3 of the 4 cities, the extremely hot summer of 2003. Record linkage of individual data from different sources offered the opportunity to exploit individual information that is rarely available in other European countries.

Some limitations must also be taken into account. The apparent temperature–mortality curves for the different cities are not identical, and these differences posed the problem of how to combine diverse information in a meta-analysis. Expressing the risk estimates as odds ratios of death resulting from high temperatures on days with temperatures of 30°C versus days with 20°C enables a straightforward synthesis of all the information available. Sensitivity analyses (results not shown) were performed varying the range from 20°–30°C, but no differences were noted from an effect-modification point of view. However, the heterogeneity among the curves makes the risk estimates more variable and limits the power when identifying specific effect modifiers.

The variables on clinical conditions are based on hospital admissions and suffer from the limits of accuracy of the source used.41 Additional data could be useful to better define chronic susceptibility (individual habits, smoking status, obesity, and so on) and acute susceptibility (place of residence, assistance received, and so on). Such information is not available from current databases. In particular, further work is needed to investigate the clinical conditions that characterized the subjects in the few weeks before death.

In conclusion, increased public awareness of the health hazards from ambient temperature and regional- or city-specific programs to prevent heat-related deaths in the elderly are public health priorities in Europe. The elderly, women, widows/widowers, and subjects with psychiatric disorders, depression, heart conduction disorders, and previous stroke have been identified as especially vulnerable during extremely hot days. The findings can help focus community and individual prevention programs, as well as responses to heat emergencies, so that associated morbidity and mortality can be prevented. Supplying adequate temperature comfort in hospitals and nursing homes seems to be an immediate and simple measure against the health effects of heat.

Back to Top | Article Outline


1. Basu R, Samet JM. Relation between elevated ambient temperature and mortality: a review of the epidemiologic evidence. Epidemiol Rev. 2002;24:190–202.

2. Semenza JC, Rubin CH, Falter KH, et al. Heat-related deaths during the July 1995 heat wave in Chicago. N Engl J Med. 1996;335:84–90.

3. Naughton MP, Henderson A, Mirabelli MC, et al. Heat-related mortality during a 1999 heat wave in Chicago. Am J Prev Med. 2002;22:221–227.

4. Rooney C, McMichael AJ, Kovats RS, et al. Excess mortality in England and Wales, and in Greater London, during the 1995 heatwave. J Epidemiol Community Health. 1998;52:482–486.

5. Braga ALF, Zanobetti A, Schwartz J. The time course of weather-related deaths. Epidemiology. 2001;12:662–667.

6. Curriero FC, Heiner KS, Samet JM, et al. Temperature and mortality in eleven cities of the eastern United States. Am J Epidemiol. 2002;155:80–87.

7. Hajat S, Kovats RS, Atkinson RW, et al. Impact of hot temperatures on death in London: a time series approach. J Epidemiol Community Health. 2002;56:367–372.

8. Alberdi JC, Diaz J, Montero JC, et al. Daily mortality in Madrid community 1986–1992: relationship with meteorological variables. Eur J Epidemiol. 1998;14:571–578.

9. Basu R, Dominici F, Samet JM. Temperature and mortality among the elderly in the United States. A comparison of epidemiologic methods. Epidemiology. 2005;16:58–66.

10. O'Neill MS, Zanobetti A, Schwartz J. Modifiers of the temperature and mortality associations in seven US cities. Am J Epidemiol. 2003;157:1074–1082.

11. World Health Organization. Health aspects of air pollution with particulate matter, ozone and nitrogen dioxide. Report from WHO Working Group Meeting, Bonn, 13–15 January 2003. Available at: http://www.euro.who.int/document/e79097.pdf. Accessed January 9, 2006.

12. Vanhems P, Gambotti L, Fabry J. Excess rate of in-hospital death in Lyons, France, during the August 2003 heat wave. N Engl J Med. 2003;349:2077–2078.

13. Schwartz J. Who is sensitive to extreme of temperature—a case-only analysis. Epidemiology. 2005;16:67–72.

14. Michelozzi P, de'Donato F, Accetta G, et al. Impact of heat waves on mortality—Rome, Italy, June–August 2003. MMWR Morb Mortal Wkly Rep. 2004;53:369–371.

15. Dhainaut JF, Claessens YE, Ginsburg C, et al. Unprecedented heat-related deaths during the 2003 heat wave in Paris: consequences on emergency departments. Crit Care. 2004;8:1–2.

16. Maclure M. The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol. 1991;133:144–153.

17. Bateson TF, Schwartz J. Control for seasonal variation and time trend in case-crossover studies of acute effects of environmental exposures. Epidemiology. 1999;10:539–544.

18. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27.

19. Kalkstein LS, Valimont KM. An evaluation of summer discomfort in the United States using a relative climatological index. Bull Am Meteorol Soc. 1986;67:842–848.

20. Katsouyanni K, Touloumi G, Spix C, et al. Short term effects of ambient sulphur dioxide and particulate matter on mortality in 12 European cities: results from time series data from the APHEA project. Air Pollution and Health: a European Approach. BMJ. 1997;314:1658–1663.

21. Samet JM, Dominici F, Curriero FC, et al. Fine particulate air pollution and mortality in 20 US cities, 1987–1994. N Engl J Med. 2000;343:1742–1749.

22. Biggeri A, Bellini P, Terracini B, et al. Metanalisi italiana degli studi sugli effetti a breve termine dell'inquinamento atmosferico 1996–2002. Epidemiol Prev. 2004;28:1–100.

23. Levy D, Lumley T, Sheppard L, et al. Referent selection in case-crossover analyses of acute health effects of air pollution. Epidemiology. 2001;12:186–192.

24. van Houwelingen HC, Arends LR, Stijnen T. Advanced methods in meta-analysis: multivariate approach and meta-regression. Stat Med. 2002;21:589–624.

25. Künzli N, Schindler C. A call for reporting the relevant exposure term in air pollution case-crossover studies. J Epidemiol Community Health. 2005;59:527–530.

26. Basu R, Samet JM. An exposure assessment study of ambient heat exposure in an elderly population in Baltimore, Maryland. Environ Health Perspect. 2002;110:1219–1224.

27. Diaz J, Garcia R, Velazquez de Castro F, et al. Effects of extremely hot days on people older than 65 years in Seville (Spain) from 1986 to 1997. Int J Biometeorol. 2002;46:145–149.

28. Davis RE, Knappenberger PC, Michaels PJ, et al. Changing heat-related mortality in the United States. Environ Health Perspect. 2003;111:1712–1718.

29. Rajpal RC, Weisskopf MG, Rumm PD, et al. Wisconsin, July 1999 heat wave: an epidemiologic assessment. WMJ. 2000;99:41–44.

30. Worfolk JB. Heat waves: their impact on the health of elders. Geriatr Nurs. 2000;21:70–77.

31. Foster KG, Ellis FP, Dore C, et al. Sweat responses in the aged. Age ageing. 1976;5:91–101.

32. Kenney WL, Hodgson JL. Heat tolerance, thermoregulation and ageing. Sports Med. 1987;4:446–456.

33. Donoghue ER, Kalelkar MB, Boehmer MA, et al. Heat-related mortality—Chicago, July 1995. MMWR Morb Mortal Wkly Rep. 1995;53:577–579.

34. Ellis FP, Nelson F, Pincus L. Mortality during heat waves in New York City July, 1972 and August and September, 1973. Environ Res. 1975;10:1–13.

35. Ellis FP, Nelson F. Mortality in the elderly in a heat wave in New York City, August 1975. Environ Res. 1978;15:504–512.

36. Vassallo M, Gera KN, Allen S. Factors associated with high risk of marginal hyperthermia in elderly patients living in an institution. Postgrad Med J. 1995;71:213–216.

37. Lye M, Kamal A. Effects of a heatwave on mortality-rates in elderly inpatients. Lancet. 1977;1:529–531.

38. Bark N. Deaths of psychiatric patients during heat waves. Psychiatr Serv. 1998;49:1088–1090.

39. Marmor M. Heat wave mortality in nursing homes. Environ Res. 1978;17:102–105.

40. Michelozzi P, de'Donato FK, Bisanti L, et al. The impact of the summer 2003 heat waves on mortality in four Italian cities. Euro Surveill. 2005;10 [Epub ahead of print].

41. Jollis JG, Ancukiewicz M, DeLong ER, et al. Discordance of databases designed for claims payment versus clinical information systems. Implications for outcomes research. Ann Intern Med. 1993;119:844–850.

Cited By:

This article has been cited 7 time(s).

Characterizing Temperature and Mortality in Nine California Counties
Basu, R; Feng, W; Ostro, BD
Epidemiology, 19(1): 138-145.
PDF (335) | CrossRef
Weather-Related Mortality: How Heat, Cold, and Heat Waves Affect Mortality in the United States
Anderson, BG; Bell, ML
Epidemiology, 20(2): 205-213.
PDF (896) | CrossRef
Temperature and Mortality in Nine US Cities
Zanobetti, A; Schwartz, J
Epidemiology, 19(4): 563-570.
PDF (639) | CrossRef
Particulate Matter and Daily Mortality: A Case-Crossover Analysis of Individual Effect Modifiers
on behalf of the SISTI Group, ; Mallone, S; Miglio, R; Pandolfi, P; Rognoni, M; Serinelli, M; Tessari, R; Vigotti, M; Perucci, CA; Forastiere, F; Stafoggia, M; Berti, G; Bisanti, L; Cernigliaro, A; Chiusolo, M
Epidemiology, 19(4): 571-580.
PDF (492) | CrossRef
Summer Temperature-related Mortality: Effect Modification by Previous Winter Mortality
Stafoggia, M; Forastiere, F; Michelozzi, P; Perucci, CA
Epidemiology, 20(4): 575-583.
PDF (467) | CrossRef
Heat Effects on Mortality in 15 European Cities
Baccini, M; Biggeri, A; Accetta, G; Kosatsky, T; Katsouyanni, K; Analitis, A; Anderson, HR; Bisanti, L; D'Ippoliti, D; Danova, J; Forsberg, B; Medina, S; Paldy, A; Rabczenko, D; Schindler, C; Michelozzi, P
Epidemiology, 19(5): 711-719.
PDF (1184) | CrossRef
Reported Ill Effects and Perceptions of the Dangers of Passive Smoking Among Indoor and Outdoor Workers in Bars in Ibadan, Nigeria
Onigbogi, OO; Akinyemi, OO
Epidemiology, 19(6): S245.
Back to Top | Article Outline

© 2006 Lippincott Williams & Wilkins, Inc.

Twitter  Facebook 


Article Tools



Article Level Metrics