ORIGINAL PAPER

Shifts in the seasonal distribution of deaths in Australia, 1968–2007

Charmian M. Bennett & Keith B. G. Dear &

Anthony J. McMichael

Received: 2 December 2012 /Revised: 7 March 2013 /Accepted: 9 March 2013# ISB 2013

Abstract Studies in temperate countries have shown thatboth hot weather in summer and cold weather in winterincrease short-term (daily) mortality. The gradual warming,decade on decade, that Australia has experienced since the1960s, might therefore be expected to have differentiallyaffected mortality in the two seasons, and thus indicate anearly impact of climate change on human health. Failure todetect such a signal would challenge the widespread assump-tion that the effect of weather on mortality implies a similareffect of a change from the present to projected future climate.We examine the ratio of summer to winter deaths against abackground of rising average annual temperatures over fourdecades: the ratio has increased from 0.71 to 0.86 since 1968.The same trend, albeit of varying strength, is evident in allstates of Australia, in four age groups (aged 55 years andabove) and in both sexes. Analysis of cause-specific mortalitysuggests that the change has so far been driven more byreduced winter mortality than by increased summer mortality.Furthermore, comparisons of this seasonal mortality ratiocalculated in the warmest subsets of seasons in each decade,with that calculated in the coldest seasons, show that particu-larly warm annual conditions, which mimic the expectedtemperatures of future climate change, increase the likelihoodof higher ratios (approaching 1:1). Overall, our results indicatethat gradual climate change, as well as short-term weathervariations, affect patterns of mortality.

Keywords Temperature . Mortality . Climate change .

Season . Poisson modelling . Mortality ratio

Introduction

Climate change-related increases in temperature are likely tohave significant impacts on health. The adverse impacts oftemperature extremes (‘heatwaves’ and ‘cold snaps’) on mor-tality are already well known, with distinct increases in mortal-ity rates (Gosling et al. 2009; Basu and Samet 2002; Bambricket al. 2008) (Fig. 1). However there has been little opportunityto examine how longer-term, more subtle changes in climate,such as the observed trend (over decades) of increasinglywarmer summers and warmer winters in Australia, might affectthe seasonal distribution of mortality. Our aim was to determinewhether this multi-decadal warming trend was associated witha concurrent shift in the seasonal distribution of all-cause andcause-specific mortality across Australia since 1968.

It is well established that the relationship between weatherand mortality differs by season (Davis et al. 2004; Guest et al.1999; McGregor et al. 2004; Gemmell et al. 2000; Lerchl1998; Wilkinson et al. 2004; Davis et al. 2003b), though moststudies of seasonal changes have focussed only on the effectsof cold on mortality. Few studies have considered trends inthis seasonal variation. Lerchl (1998) found that while lowtemperatures had a stronger impact than heat on mortality inGermany, the impact of cold has been declining over the last~50 years, probably related to increased use of central heatingand health system improvements. Carson and colleagues(Carson et al. 2006) found that the ratio of winter:non-winterdeaths in London shifted over different time periods since1900, but with no consistent trend.

In many parts of the world, including Australia, there is aclear seasonal distribution in mortality, typically entailinghigher mortality rates in winter and summer than in spring

C. M. Bennett (*) :K. B. G. Dear :A. J. McMichaelNational Centre for Epidemiology and Population Health, TheAustralian National University, Canberra, ACT 0200, Australiae-mail: charmian.bennett@anu.edu.au

K. B. G. Deare-mail: keithdear4@gmail.com

A. J. McMichaele-mail: tony.mcmichael@anu.edu.au

Int J BiometeorolDOI 10.1007/s00484-013-0663-x

and autumn. On a daily basis, there is a stronger associationbetween mortality and heat than between mortality and cold,as evidenced by large surges in mortality associated withextremely hot days and extended hot periods (‘heatwaves’)that are not matched by equivalent surges in mortality on verycold days (Kalkstein and Greene 1997). Nevertheless, in mosttemperate climates, winter mortality rates are typically signifi-cantly higher than those in summer due to the winter seasonalimpacts of infectious diseases (especially influenza) and theexacerbation of respiratory and cardiovascular diseases (Daviset al. 2004). In Australia, a clear seasonal signal is observed inall-cause mortality, with higher mortality in winter than insummer (Guest et al. 1999). Total summer deaths from 1968to 2007 in Australia were 80.0 % of the number of winterdeaths. The greater number of winter season deaths is largelyexplained by infectious disease transmission peaks during win-ter and the exacerbation of chronic diseases, especially cardio-vascular and respiratory conditions (Cameron et al. 1985).

Climate change projections consistently indicate that aver-age temperatures in many places will rise, resulting in bothwarmer summers (more very hot days) and warmer winters(fewer very cold days). The frequency of temperature extremesin Australia appears to be changing in a similar manner, withmore extremely hot days and fewer extremely cold days ob-served every decade since 1960. The average number of recordhot days in summer has increased every decade, from ~10 daysper summer in the decade 1960–1969 to ~23 days per summerin the decade 2000–2009, whilst over the same period, thenumber of record cold days in winter has decreased everydecade, from ~21 days in the decade 1960–1969 to ~10 daysin the decade 2000–2009 (Bureau of Meteorology and CSIRO2010). Therefore, it is expected that the dominance of wintermortality will decline as summer mortality rates rise.

A shift in the seasonal distribution of deaths is consistentwith the hypothesised effect of a warming climate, andsupport for this hypothesis can be strengthened by exploringvariations in the trend by place, person and disease class, tosee whether these variations are also as might be expected.Our research examined the seasonal distribution of mortal-ity, including by age, sex, cause of death and location, tovalidate this hypothesis in the Australian context.

Methods

We examined the ratio of summer-to-winter deaths from1968 to 2008, for the whole of Australia, using all-causeand cause-specific deaths in those aged 55 years and above.

Data

All-cause mortality data for those aged 55 and over wereobtained from the Australian Bureau of Statistics for alleight states and territories. Four age groups were used inthe analysis: 55–64 years, 65–74 years, 75–84 years, and85+ years.

Cause-specific mortality was identified using the firstthree digits of the International Classification of Diseases(Australian Modification) (ICD-AM) mortality codes. Thestudy period spanned versions 8, 9 and 10 of the ICD-AM,so diagnosis codes were matched across ICD versions toensure consistency. The ICD codes used in this study arelisted in Table 1. Respiratory and cardiovascular diseaseshave been widely linked to seasonal variations in tempera-ture as well as to shorter-term temperature extremes (such asheat waves and winter storms) (Analitis et al. 2008; Basu

Fig. 1 Association betweentemperature (3-day average)and all-cause mortality in adultsaged 45+ in Brisbane, 1990–2007. The curve is a 6dfgeneralised additive model,adjusted for long-term trend andannual cycles. Relative risk isexpressed relative to theminimum, which is at about21 °C

Int J Biometeorol

and Samet 2002; Gosling et al. 2009). Renal disease wasalso included as recent studies have suggested that heat-related dehydration and hyperthermia are associated withrenal dysfunction, especially in the elderly (Nitschke et al.2011; Hansen et al. 2008).

‘Winter’ was defined as the three coldest months (June,July and August), and ‘Summer’ as the three warmestmonths (December, January and February). The availabledata included 40 winters from 1968 to 2007, and 39 sum-mers from 1968/69 to 2006/07.

Analyses were performed using Stata12SE (Statacorp,College Station, TX).

Regression analysis

Poisson regression models of the number of deaths perseason were used. The ratio of summer to winter mortalitywas modelled as a 2-level (1df) covariate, and the trend inthis ratio over time was captured as the interaction of thisfactor with ‘year’, treated as continuous. Year k was codedas the calendar year (k=1968…2007) for the 40 winterseasons (mortality in June, July and August combined),and k=1968.5…2006.5 for the 39 intervening summer sea-sons. For example, total mortality in the summer seasonDecember 1968, January 1969 and February 1969 wasconsidered to occur at time year=1968.5.

The model was adjusted for age group (four levels), sex,and location (eight states and territories).

We defined “demographic setting” to represent the 64possible combinations of location, age and sex. Given ademographic setting i, season j (1=summer and 2=winter),and year k (1968, 1968.5, … , 2006.5, 2007), our basemodel for seasonal deaths Yijk was

logE Yijk� � ¼ aij þ bi � k þ cj � k

The intercepts aij account for population denominators,but also allow for differential overall mortality by demo-graphic setting (i) in each season (j). Thus this term implic-itly includes the main effects of age, sex, location and

season and also all pairwise and higher interactions betweenthese four factors. The bi terms represent linear setting-specific trends in mortality, to capture demographic driftsand general trends in society and local health care. The cjterms capture any overall differences in these trends be-tween the two seasons, and are the parameters of principalinterest in our analysis. To avoid overparameterisation, weapplied the constraint c2=0. In this base model, the param-eter c1 then expresses an annual increase in log deaths insummer only, above any setting-specific year-round trends,and this extra rate of change is assumed, for now, to be thesame for all subpopulations.

Analyses were then stratified by location, age group, sexand cause of death, by augmenting the base model withadditional interaction terms. The effect of this was to replacethe single parameter c1 with sets of parameters for differentages, sexes, locations, and disease classes.

Within-decade comparison

A second, distinct, analysis was performed to assess theinfluence of short-term, seasonal, variations in temperatureon mortality ratios, independent of the mortality ratio trendover time. This analysis examined whether individual hotand cold seasons were associated with changes in the sea-sonal mortality ratio, and to estimate the relative importanceof more extreme seasonal temperatures, given that futureclimate change in Australia is expected to bring a morevariable climate (CSIRO and Bureau of Meterology 2007).This analysis also provided a test of the validity of ourinterpretation of the 40-year trends examined earlier. If, asexpected, a trend emerges over the four decades studied, thiscould be due to climate change or to some other unidentifiedconfounding cause. In the latter case, within-decade com-parisons of hot seasons should show similar ratios to com-parisons of cold seasons.

Mean seasonal temperature for the 40 winters and 39summers studied was calculated using data from theAustralian Bureau of Meteorology. Summers (DJF) wereassigned the December year+0.5, and the study period wasthen divided into four decades (1968–1977.5, 1978–1987.5,1988–1997.5, and 1998–2007). The seasonal mean tempera-tures were ranked to identify the hottest and coolest summers,and the warmest and coldest winters, in each decade.

The mortality ratio H1 associated with the hottest twoseasons was first calculated as the four-decade averageof the mean mortality during the single hottest summerdivided by that in the single warmest winter in eachdecade:

H1 ¼ 1

4

S1;1W1;1

þ S2;1W2;1

þ S3;1W3;1

þ S4;1W4;1

� �

Table 1 International Classification of Disease—Australian Modifi-cation (ICD-AM) mortality codes used to identify deaths from respi-ratory, cardiovascular and renal causes.

Cause of death ICD codes used

ICD 8 (1968–1978)and ICD 9 (1979–1998)

ICD 10 (1999–2007)

Respiratorydiseases

460–519, 786 J00-J99

Cardiovasculardiseases

390–459, 785 G45-G46, I00-I99,R00

Renal diseases 580–599 N00-N99

Int J Biometeorol

Here Si,1 is the hottest summer and Wi,1 is the warmestwinter in the ith decade.

The mortality ratio H2 was then calculated using meanmortality during the hottest two summers and two warmestwinters, then the hottest three summers and warmest threewinters, and so on until the mortality ratio for all ten sum-mers and ten winters per decade was calculated.

Hk ¼ 1

4

X4i¼1

Pkj¼1 SijPkj¼1 Wij

!

where k=1…10 points on graph (Fig. 3); i=1…4 decades;j=1…10 ranked seasons; Sij is the jth hottest summer in the ithdecade; and Wij is the jth warmest winter in the ith decade.

These steps were repeated for the lowest mean tempera-tures, using the coldest winter and coolest summer, then thecoldest two winters and coolest two summers, and so on,resulting in indices Ck. Necessarily, H10=C10 since all sea-sons are then included.

We could use only nine summers in the last decadebecause the potential tenth summer, December 2007 toFebruary 2008, extended beyond the available data. Forthe last point on the graph the contribution from decadefour was therefore taken to be the average mortality overnine summers divided by the average mortality over tenwinters.

Results

There has been a gradual increase in the number of deathseach year since 1968 (Fig. 2), which reflects populationgrowth over time. Seasonal deaths in winter are more

variable than in summer, probably due to the greater impactof communicable disease outbreaks in winter.

The seasonal mortality ratio of summer-to-winter all-cause deaths increased over the study period, from 0.709in 1968 to 0.859 in 2006 (data from summer 2007 was notavailable). In 1968, there were approximately 71 deaths insummer for every 100 deaths in winter. By 2007, this hadrisen to approximately 86 deaths in summer for every 100deaths in winter.

Our modelling estimated that the summer:winter mortal-ity ratio increased by a factor of 1.00339 per year—anannual rise of 0.339 % (95 % CI 0.31–0.36 %, P<0.001;Table 2).

There was a consistent, statistically significant trend ofannual increases in mortality ratios across all of the stratifiedanalyses (by sex, age and location, as shown in Table 2). Themodelled annual change in all-cause mortality for males wasalmost double that for females (P<0.001). The ratio increasedby approximately 0.4 % per annum for ages 55–74 years, witha smaller annual increase for age 75–84 years and markedlysmaller annual increase for age 85+. There were significantdifferences between states and territories across Australia,ranging from 0.42 % per annum in Queensland to 0.17 %per annum in South Australia (P<0.001).

The trend in the mortality ratio also varied strongly bycause of death (Table 3). Themortality ratio for cardiovasculardeaths was estimated to increase by 0.26 % per year (95 % CI0.22–0.29 %, P<0.001), while the trend in the ratio for respi-ratory deaths was estimated to increase by 0.65 % per year(95 % CI 0.55–0.76 %, P<0.001). There was no significantchange in the ratio for renal causes of death.

Figure 3 indicates that short-term, seasonal changes in tem-perature were related to the summer:winter mortality ratio, and

Fig. 2 Seasonal deaths over 40winters and 39 summers. Thesummer:winter ratio wascalculated as the number ofdeaths in each summer dividedby the average number ofdeaths in the two adjacentwinters, and smoothed using amoving average (dotted line). Alog-linear trend is also shown(dashed line)

Int J Biometeorol

that this association was independent of time. In this analysis,the seasons were ranked according to the mean seasonal tem-perature, not chronologically, and Fig. 3 clearly shows thatwarmer seasons were associated with higher summer:wintermortality ratios than cooler seasons. The final points coincidebecause the warmest 10 years in each decade are, necessarily,identical to the coolest 10 years in each decade.

Discussion

The gradual increase in the number of deaths each year since1968 (as shown in Fig. 1) reflects population growth overtime. Overall, mortality rates are highest during the wintermonths, largely as a result of infectious diseases (includinginfluenza) and the exacerbation of chronic respiratory andcardiovascular diseases.

Our analyses showed that there has been a statisticallysignificant increase in the ratio of summer-to-winter mortality

over the last four decades.We found that the ratio significantlyincreased in all regions of Australia, in both sexes, in all adultage groups and for three causes of death that have previouslybeen shown to demonstrate a seasonal distribution (Analitis etal. 2008; Basu and Samet 2002; Davis et al. 2004).

The trends in the summer:winter mortality ratios variedby sex and age. The trend was noticeably higher for malesthan females. The change over time in mortality ratio wassimilar for adults aged 55–64 years and 65–74 years, butsmaller in the older age groups. This muted seasonal re-sponse with increasing age probably reflects a mix of biol-ogy, residential environment, occupational environment andbehavioural patterns that modify exposure to temperatureextremes. For example, older people may spend much oftheir time indoors in climate-controlled environments (pri-vate homes and residential care facilities), and thus be lesslikely to experience extremes of temperature. Further, com-municable diseases and chronic illnesses probably dominatemortality trends among older age groups, so that the

Table 2 Modelled annual change in summer:winter all-cause mortality ratio, including stratified analyses

Overall summer:winter mortality ratio

Modelled annual changein summer:winterall-cause mortality ratio (%)

95 %Confidenceinterval

P-value Likelihood ratiotest for stratifiedanalysis (P-value)a

All-cause deaths 0.800 : 1 0.339 0.314, 0.363 <0.001

Male 0.803 : 1 0.428 0.394, 0.463 <0.001 <0.001Female 0.796 : 1 0.242 0.206, 0.277 <0.001

Age 55–64 years 0.868 : 1 0.410 0.346, 0.474 <0.001 <0.001Age 65–74 years 0.834 : 1 0.415 0.367, 0.463 <0.001

Age 75–84 years 0.790 : 1 0.343 0.301, 0.384 <0.001

Age 85+ years 0.740 : 1 0.198 0.147, 0.250 <0.001

New South Wales 0.781 : 1 0.332 0.291, 0.372 <0.001 <0.001Victoria 0.827 : 1 0.400 0.352, 0.448 <0.001

Queensland 0.794 : 1 0.416 0.355, 0.477 <0.001

South Australia 0.800 : 1 0.167 0.085, 0.249 <0.001

Western Australia 0.796 : 1 0.236 0.146, 0.326 <0.001

Tasmania 0.813 : 1 0.250 0.108, 0.393 0.001

Northern Territory 0.865 : 1 0.319 −0.109, 0.746 0.145

Australian Capital Territory 0.783 : 1 0.302 0.005, 0.599 0.046

a Testing Ho : same slope in all strata

Table 3 Modelled annualchange in summer:winter dis-ease-specific mortality ratios

Overall summer:winter mortalityratio (3dp)

Modelled annual changein summer: wintermortality ratio (%, 3dp)

95 % Confidenceinterval (%, 3dp)

P-value

All-cause 0.800 : 1 0.339 0.314, 0.363 <0.001

Cardiovascular deaths 0.776 : 1 0.256 0.221, 0.292 <0.001

Respiratory deaths 0.598 : 1 0.653 0.548, 0.758 <0.001

Renal deaths 0.772 : 1 0.165 −0.029, 0.360 0.096

Int J Biometeorol

association between mortality and temperature appears todecline with increasing age.

The trend in the mortality ratio varied considerably be-tween states and territories. The greatest increases wereobserved in the eastern and southeastern regions, wherethe majority of Australia’s population reside in a numberof large cities. Smaller increases in the mortality ratio wereobserved in South Australia and Western Australia. Theseregions typically experience more frequent heat extremesand a generally hotter climate, so the smaller increases in themortality ratio may reflect some degree of physiological andcultural acclimatisation in these populations. Similar varia-tions in temperature–mortality relationships have been iden-tified previously in other countries, and are thought to berelated to population acclimatisation to the local climate,with mortality affected most strongly by the unusualnessof the temperature extreme (hot or cold), rather than itsabsolute value (Medina-Ramon and Schwartz 2007;Analitis et al. 2008; McMichael et al. 2008). The trends inthe Australian Capital Territory and Northern Territory weresimilar to those in the six states of Australia, but were lessstatistically significant due to their smaller populations.

Mortality statistics associated with extremes of temperaturemay reflect in part the premature deaths of those who werealready likely to die in the near future, rather than the deaths ofotherwise healthy individuals (known as mortality displace-ment, or ‘harvesting’). Studies have shown that mortalitydisplacement, when it occurs, is most likely associated withshort-term spells of temperature extremes (particularly heat),and that deaths are displaced by only a few days to weeks(Braga et al. 2001; Hajat et al. 2005; Kysely 2004). Evidencefor the seasonal displacement of deaths is inconsistent(Rocklov et al. 2009; Toulemon and Barbieri 2008; Ha et al.2011; Basu and Malig 2011), and so we assumed that the

deaths in each season studied were independent of the deathsin the season before.

The all-cause summer:winter mortality ratio was found tobe higher (closer to one) than for any of the three disease-specific ratios we examined. It is likely that this reflects theinfluence of other causes of death, such as cancers, that lackseasonal variation. We examined respiratory, cardiovascularand renal causes of death, which have been previously shownto vary by season (Analitis et al. 2008; Basu and Samet 2002;Davis et al. 2004) and thus are also likely to be affected bychanges in temperature and temperature distribution associat-ed with future climate change. Our results showed that themortality ratios for all three causes of death are increasing.Respiratory deaths in particular are highest in winter, so theoverall mortality ratio is lower (more extreme) than for theother causes of death examined. However, the ratio for respi-ratory deaths was also shown to be increasing the fastest,indicating that considerably fewer winter respiratory deathsand/or more summer deaths are occurring each year.

We had hypothesised that rising temperatures over time inAustralia would be reflected in a trend towards highersummer:winter mortality ratios, and our results support this.This trend could also be associated with changes in publichealth care, treatment protocols and other unknown demo-graphic changes in the population. However, our analysis ofwithin-decade variations in the summer:winter mortality ratioindicated that changes in the mortality ratio are independentlylinked to changes in temperature. It is therefore reasonable toconclude that this 40-year trend is related to changes in tem-perature over time.

Based on climate change knowledge, evidence and pro-jections, we would assume that the summer:winter mortalityratio will increase as the climate warms in the future. It isunclear whether this will be due to an increase in the number

.75

.76

.77

.78

.79

.8

Rat

io o

f sum

mer

:win

ter

mor

talit

y1 2 3 4 5 6 7 8 9 10

Number of summer : winter comparisons per decade

Hottest summers:warmest wintersCoolest summers:coldest winters

Fig. 3 The summer:wintermortality ratio in seasonsselected within each decade.Using only cool seasons (bothsummers and winters) givessmaller ratios (diamonds),while selecting warm seasonsgives larger ratios (squares) andmimics longer-term climatechange. See text for details

Int J Biometeorol

of summer (heat-related) deaths, or a decline in winter (cold -related) deaths, or some combination of both. In the shortterm, an increase in the summer:winter mortality ratio couldbe interpreted as a ‘positive’/beneficial impact, meaning fewerdeaths occurring each winter. However, in the long-term, withclimate change expected to increase average ambient temper-atures as well as bring more frequent extreme heat events insummer, the resultant absolute increase in summer deaths willbecome an increasingly detrimental outcome.

The overall impact of climate change on absolute seasonalmortality rates is uncertain (Gosling et al. 2009). Some arguethat any increase in heat-related (summer) mortality will beoffset by a decrease in cold-related (winter) mortality, thushaving little overall impact on total annual mortality (Davis etal. 2003a, 2004). For five major coastal Australian cities,Guest and colleagues (1999) predicted that future excessmortality for the hottest days in summer would be greater thanthat for the coldest days in winter, but overall, they predictedan 8–12 % reduction in mortality in 2030 (Guest et al. 1999).Others argue that the magnitude of increase in heat-related(summer) mortality, associated with higher ambient tempera-tures as well as more frequent periods of extreme heat withfuture climate change, will outweigh any modest decreases incold-related (winter) mortality associated with fewer extreme-ly cold days and, as such, climate change is likely to increaseannual mortality markedly (Kalkstein and Greene 1997;Nicholls 2009; McMichael et al. 2003; Medina-Ramon andSchwartz 2007). In all of these scenarios, the true impact ofseasonal shifts in mortality is likely to vary strongly by loca-tion, associated with the demographic profile, acclimatisationand adaptive capacity of the population. Only time will revealexactly how climate change-related shifts in temperature dis-tributions will affect mortality trends. In the meantime, it isprudent to implement public health strategies that consider thepossible impacts of both hot and cold temperatures on localmortality trends.

Furthermore, Australia is a relatively affluent country thatis used to hot summers. That a shift in seasonal mortality isevident here could indicate that even greater shifts are likelyin other ‘cooler’ countries, where high summer temperatures(>35 °C) are unfamiliar.

Conclusion

There has been a statistically significant increase in the ratioof summer:winter deaths over 40 years in Australia, evidentacross different age groups, sexes, locations and causes ofdeath. Short-term (within decade) comparisons of mortalityin seasons selected by temperature, designed to mimic long-term warming, showed a similar pattern of higher ratios inwarmer conditions. Taken together, these results suggestthat relatively gradual changes in climate, and not only

short-term weather fluctuations, affect mortality rates, andtherefore that the influence of climate change is alreadydetectable in the recent Australian mortality record.

Acknowledgements C.M.B. and A.J.M. were supported by NationalHealth and Medical Research Council Australia Fellowship 0418141.This work was also supported by National Health and Medical Re-search Council Project Grant 585408.

References

Analitis A, Katsouyanni K, Biggeri A, Baccini M, Forsberg B, BisantiL, Kirchmayer U, Ballester F, Cadum E, Goodman P, Hojs A,Sunyer J, Tittanen P, Michelozzi P (2008) Effects of cold weatheron mortality: results from 15 European Cities Within the PHEWEProject. Am J Epidemiol 168:1397–1408

Bambrick H, Dear K, Woodruff R, Hanigan I, McMichael A (2008)The impacts of climate change on three health outcomes:temperature-related mortality and hospitalisations, salmonellosisand other bacterial gastroenteritis, and population at risk fromdengue. Garnaut Climate Change Review, Canberra

Basu R, Malig B (2011) High ambient temperature and mortality inCalifornia: Exploring the roles of age, disease and mortalitydisplacement. Environ Res 111:1286–1292

Basu R, Samet JM (2002) Relation between elevated ambient temper-ature and mortality: a review of the epidemiological evidence.Epidemiol Rev 24:190–202

Braga ALF, Zanobetti A, Schwartz J (2001) The time course ofweather-related deaths. Epidemiology 12:662–667

Bureau of Meteorology, CSIRO (2010) State of the Climate. AustralianGovernment Bureau of Meteorology & Commonwealth Scientificand Industrial Research Organisation, Melbourne

Cameron AS, Roder DM, Esterman AJ, Moore BW (1985) Mortalityfrom influenza and allied infections in South Australia during1968–1981. Med J Aust 142:14–17

Carson C, Hajat S, Armstrong B, Wilkinson P (2006) Declining vul-nerability to temperature-related mortality in London over the20th Century. Am J Epidemiol 164:77–84

CSIRO and Bureau of Meterology (2007) Climate change in Australia:Technical Report 2007. Melbourne: CSIRO, BoM. http://www.climatechangeinaustralia.gov.au.

Davis RE, Knappenberger PC, Michaels PJ, Novicoff WM (2003a)Changing heat-related mortality in the United States. EnvironHealth Perspect 111:1712–1718

Davis RE, Knappenberger PC, Michaels PJ, Novicoff WM (2004)Seasonality of climate-human mortality relationships in US citiesand impacts of climate change. Clim Res 26:61–76

Davis RE, Knappenberger PC, Novicoff WM, Michaels PJ (2003b)Decadal changes in summer mortality in US cities. Int JBiometeorol 47:166–175

Gemmell I, McLoone P, Boddy FA, Dickinson GJ, Watt GCM (2000)Seasonal variation in mortality in Scotland. Int J Epidemiol29:274–279

Gosling SN, Lowe JA, McGregor GR, Pelling M, Malamud BD (2009)Associations between elevated atmospheric temperature and hu-man mortality: a critical review of the literature. Clim Chang92:299–341

Guest CS, Willson K, Woodward AJ, Hennessy K, Kalkstein LS,Skinner C, McMichael AJ (1999) Climate and mortality inAustralia: retrospective study, 1979–1990, and predicted impactsin five major cities in 2030. Clim Res 13:1–15

Int J Biometeorol

Ha J, Kim H, Hajat S (2011) Effect of previous-winter mortality on theassociation between summer temperature and mortality in SouthKorea. Environ Health Perspect 119:542–546

Hajat S, Armstrong BG, Gouveia N, Wilkinson P (2005) Mortalitydisplacement of heat-related deaths: a comparison of Delhi, SaoPaulo, and London. Epidemiology 16:613–620

Hansen AL, Bi P, Ryan P, Nitschke M, Pisaniello D, Tucker G (2008)The effect of heat waves on hospital admissions for renal diseasein a temperate city of Australia. Int J Epidemiol 37:1359–1365

Kalkstein LS, Greene JS (1997) An evaluation of climate/mortalityrelationships in large U.S. cities and the possible impacts of aclimate change. Environ Health Perspect 105:84–93

Kysely J (2004) Mortality and displaced mortality during heat waves inthe Czech Republic. Int J Biometeorol 49:91–97

Lerchl A (1998) Changes in the seasonality of mortality in Germanyfrom 1946 to 1995: the role of temperature. Int J Biometeorol42:84–88

McGregor GR, Watkin HA, Cox M (2004) Relationships between theseasonality of temperature and ischaemic heart disease mortality:implications for climate based health forecasting. Clim Res25:253–263

McMichael AJ, Wilkinson P, Kovats RS, Pattenden S, Hajat S,Armstrong B, Vajanapoom N, Niciu EM, Mahomed H, KingkeowC, Kosnik M, O’Neill MS, Romieu I, Ramirez-Aguilar M, BarretoML, Gouveia N, Nikiforov B (2008) International study of

temperature, heat and urban mortality: the ‘ISOTHURM’ project.Int J Epidemiol 37:1121–1131

McMichael AJ, Woodruff RE, Whetton P, Hennessy K, Nicholls N,Hales S, Woodward A, Kjellstrom T (2003) Human health andclimate change in Oceania: a risk assessment 2002.Commonwealth Department of Health and Ageing, Canberra

Medina-Ramon M, Schwartz J (2007) Temperature, temperature ex-tremes, and mortality; a study of acclimatisation and effect mod-ification in 50 US cities. Occup Environ Med 64:827–833

Nicholls N (2009) Estimating changes in mortality due to climatechange. Clim Chang 97:313–320

Nitschke M, Tucker GR, Hansen AL, Williams S, Zhang Y, Bi P(2011) Impact of two recent extreme heat episodes on morbidityand mortality in Adelaide, South Australia: a case-series analysis.Environ Health 10:42

Rocklov J, Forsberg B, Meister K (2009) Winter mortality modifies theheat-mortality association the following summer. Eur Respir J33:245–251

Toulemon L, Barbieri M (2008) The mortality impact of the August2003 heat wave in France: investigating the ‘harvesting’ effectand other long-term consequences. Popul Stud 62:39–53

Wilkinson P, Pattenden S, Armstrong B, Fletcher A, Kovats RS,Mangtani P, McMichael AJ (2004) Vulnerability to winter mor-tality in elderly people in Britain: population based study. BMJ329:647

Int J Biometeorol