13 Burden of disease Pam Groenewald, Debbie Bradshaw ... Health Barometers/13... · n Dorrington R, Bradshaw D. Maternal mortality in South Africa: lessons from a case study in the

218

Section A: Burden of disease

13 Burden of disease Pam Groenewald, Debbie Bradshaw,

Candy Day, and Ria Laubscher

District-level mortality information is extremely important for health managers and programme planners in order to monitor health status, assess effectiveness of priority programmes, and identify emerging health issues and vulnerable groups. Such data can also be used to gauge inequities in health among districts. Despite the data-quality concerns, it is essential to start making use of the available data, at the same time as initiating data-improvement strategies. By assuming that the metro areas have near-complete death registration, it is possible to obtain death rates for these metros. While it is not yet possible to provide reliable mortality rates for each district, the epidemiological mortality profiles can help to measure the need for equitable resource allocation and priority setting.

This is the fourth attempt in the District Health Barometer publication to assess and compare the cause of death profiles for each of the 52 health districts in South Africa. Currently, Statistics South Africa (Stats SA) compiles cause of death statistics based on death notifications, but reports only limited information at district level. Furthermore, data-quality issues have been identified, including a high proportion of ill-defined causes, misclassification of HIV-related deaths, and poor specification of external causes of injury related deaths.a

Methodology

Data source

Unit records for the 2008–2013 mortality data were provided by Stats SA.b,c,d,e,f,g Data included age, sex, district of death and underlying cause of death coded to the International Statistical Classification of Diseases and Related Health Problems (ICD-10).h The ICD classification contains a detailed list of causes of mortality that is too extensive for public health use. For this reason the ICD codes were aggregated according to the National Burden of Disease (NBD) list,i which is a condensed list of conditions containing the most prevalent diseases across South Africa, including those of public health importance. The NBD list has recently been updated,j and differs slightly from that used for the district mortality profiles prepared for the first District Health Barometer. Vital statistics data are updated annually with late registrations. For these reasons, the data for 2008, 2009, 2010 and 2011 were re-analysed with the 2012 and 2013 data. Stillbirths were excluded from the data prior to analysis.

Aggregation of causes of death

The NBD list of causes was aggregated into three broad cause groups, namely communicable diseases together with perinatal, maternal and nutritional conditions (Comm/Mat/Peri/Nutr); non-communicable diseases (NCDs); and injuries; as indicated in the 2000 NBD studyi (see Table 1). Given the large burden caused by HIV-related deaths, which form part of the communicable disease group, these deaths were separated into a fourth group. Since many HIV deaths are misclassified to tuberculosis (TB), the TB deaths were reported with the HIV deaths.

a Bradshaw D, Pillay-van Wyk V, Laubscher R, Nojilana B, Groenewald, Nannan N. Cause of death statistics for South Africa: Challenges and possibilities for improvement. Cape Town: Medical Research Council, 2011. www.mrc.ac.za/bod/cause_death_Stats SA.pdf [accessed 30 Oct 2012].

b Statistics South Africa. Mortality and causes of death in South Africa, 2008: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2010. http://www.Stats SA.gov.za/publications/P03093/P030932008.pdf [accessed23 Sept 2014].

c Statistics South Africa. Mortality and causes of death in South Africa, 2009: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2011. http://www.Stats SA.gov.za/publications/P03093/P030932009.pdf [accessed 23 Sept 2014].

d Statistics South Africa. Mortality and causes of death in South Africa, 2010: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2013. http://www.Stats SA.gov.za/publications/P03093/P030932010.pdf [accessed 23 Sept 2014].

e Statistics South Africa. Mortality and causes of death in South Africa, 2011: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2014. http://www.Stats SA.gov.za/publications/P03093/P030932011.pdf [accessed23 Sept 2014].

f Statistics South Africa. Mortality and causes of death in South Africa, 2012: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2014. http://www.Stats SA.gov.za/publications/P03093/P030932012.pdf [accessed23 Sept 2014]

g Statistics South Africa. Mortality and causes of death in South Africa, 2013: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2015. http://www.Stats SA.gov.za/publications/P03093/P030932013.pdf [accessed23 Sept 2014]

h World Health Organization. International Statistical Classification of Diseases and Related Health Problems. 10th revision. Volume 2. 2nd ed. Geneva: World Health Organization, 2004. http://www.who.int/classifications/icd/ICD-10_2nd_ed_volume2.pdf [accessed 30 Oct 2012].

i Bradshaw D, Groenewald P, Laubscher R, et al. Initial burden of disease estimates for South Africa, 2000. Cape Town: South African Medical Research Council, 2003. www.mrc.ac.za/bod/initialbodestimates.pdf [accessed 30 Oct 2012].

j Pillay-Van Wyk V, Laubscher R, Msemburi W, Dorrington RE, Groenewald P, Vos T, et al. Second South African National Burden of Disease Study: Data cleaning, validation and SA NBD List. Cape Town: Burden of Disease Research Unit, South African Medical Research Council.

219


Table 1: Examples of causes of death in each broad cause group

Broad cause group ExamplesCommunicable diseases (excluding HIV and TB) maternal, perinatal and nutritional disorders(Comm/Mat/Peri/Nut)

Diarrhoeal diseases Meningitis & encephalitis Maternal conditions Perinatal conditions Nutritional disorders

HIV related and TB (HIV and TB) HIV related Tuberculosis

Non-communicable diseases (NCDs) Cerebrovascular disease Diabetes MellitusIschaemic heart diseaseCancer

Injuries Transport injuries Interpersonal violence

Adjustments to data

STATA 13 was used to adjust the data, firstly by redistributing deaths of unknown age and sex proportionally by known age and gender across each of the known causes of death and districts. Causes of death used as pseudonyms for AIDS, e.g. ‘retroviral disease’ or ‘immune suppression’ were combined with the HIV deaths. Deaths misclassified to ill-defined signs and symptoms (ICD chapter XVII) and other ‘garbage codes’ (such as intermediate causes of death, e.g. septicaemia; mechanisms of death, e.g. cardiac arrest; partially specified causes, e.g. cancer with unknown site of the disease; or risk factors, e.g. hypertension)k were proportionally redistributed to specified causes within each age and sex category.

Cause of death information for injuries was particularly problematic, with a very high proportion of ‘undetermined cause’ due to the manner of death (accident, homicide, suicide) not being specified on the death notification form. To accommodate a coding change implemented by Stats SA in 2007,l whereby unspecified injuries are coded to accidental injuries according to ICD-10 guidelines, injuries were redistributed using a different redistribution algorithm. This involved identifying the proportion of accidental injuries that would previously have been coded as unspecified based on 2006 data, and re-allocating these proportionally to homicide, suicide and accidental intent. In the absence of district level information, the estimated national proportions were applied to each district, based on the assumption that the change in coding was consistent across the country.

Analysis

The proportions of deaths and years of life lost (YLLs) due to the four broad cause groups, were calculated for each of the 52 districts. YLLs are a measure of premature mortality based on the age at death and therefore highlight the causes of death that should be targeted for prevention. In line with the initial South African NBD study, the highest observed national life expectancy was selected as the standard against which YLLs are calculated.m

Death registration for 2008 was reported to be 81% nationally,b but Dorrington and Bradshaw estimate that it was higher at 90%.n Registration for 2009 was reported to be 93.5% at national level,c and for 2010, 2011,e 2012f and 2013 it was reported to be 94%.g However, estimates of the completeness of death registration are not available at district level, and since variation in completeness at district level could distort death rates, these were not calculated except for the eight metros where completeness was likely to be good. The numbers of deaths, age distribution and the seasonal trends for each year were examined and compared for all districts. Metro death rates were age standardised to eliminate differences in observed mortality rates caused by differences in the age structure of the population in different areas.o Rates were calculated using the population estimates from the District Health Information Software (DHIS), based on 2002–2018 district cohort estimates developed by Statistics South Africa (2013).

k Naghavi M, Makela S, Foreman K, O’Brien J, Pourmalek F, Lozano R. Algorithms for enhancing public health utility of national causes of death data. Popul Health Metr. 2010;8:9.

l Statistics South Africa. Mortality and causes of death in South Africa, 2007: findings from death notification. Statistical Release P0309.3. Pretoria: Statistics South Africa, 2009. http://www.Stats SA.gov.za/publications/P03093/P030932007.pdf [accessed 30 Oct 2012].

m This standard is represented by a model life table, Coale and Demeny West level 26, with a life expectancy at birth of 82.5 years for Japanese females and 80 for males. YLLs are estimated for each age, sex and cause category by multiplying the observed number of deaths in each category by the expected life expectancy in each age category, implying that YLLs are greater when age at death is younger. Since people value years of life gained in the future less than years gained in the present, a 3% discount rate is applied. In contrast to the first NBD study, an age weighting function that assigns greater value to a year of life lived in the economically active age groups than it assigns to years lived in childhood or old age was not applied, in line with the latest Global Burden of Disease protocol (http://www.dcp2.org/pubs/GBD).

n Dorrington R, Bradshaw D. Maternal mortality in South Africa: lessons from a case study in the use of deaths reported by households in censuses and surveys. J Pop Research. 2011; 28:49-73.

o Ahmad OB, Boschi-Pinto C, Lopez AD, Murray CJL, Lozano R, Inoue M. Age standardisation of rates – A new WHO standard, GPE Discussion Paper Series No 31. Geneva: WHO; 2001.

District DeathDate

January February March April May June July August September October November December

eThekwini:ETH

1000

1500

2000

2500

3000

3500

Cou

nt o

f Ser

ial N

o

Monthly_trends

Year of DeathDate2008

2009

2010

2011

2012

2013

220


Results

A total of 3 281 156 deaths were reported from 2008 to 2013, with 87 672 stillbirths excluded from further analysis (Table 2). There was a decline in the total number of deaths between 2008 and 2013, in keeping with the noted decline in mortality rates.p

Between 2008 and 2009, large fluctuations were noted in the total number of deaths reported by Chris Hani (Eastern Cape (EC)), Nkangala (Mpumalanga (MP)) and ZF Mgcawu (previously Siyanda) (Northern Cape (NC)) districts; these were completely inconsistent with the trends over the previous 12 years.q We were unable to explain these inconsistencies. Between 2009 and 2010, there were large fluctuations in the total numbers of deaths reported by Nelson Mandela Bay and OR Tambo (both in the EC), Mangaung and Xhariep in the Free State (FS), Johannesburg, Sedibeng and Tshwane in Gauteng (GP), eThekwini and Harry Gwala in KwaZulu-Natal (KZN), Capricorn in Limpopo (LP) and Pixley ka Seme and ZF Mgcawu in the Northern Cape (NC). Some of these fluctuations can be explained by 6 704 deaths where the district was ‘unknown’. Similarly, in 2011 and 2012 there were a large number of deaths (23 165 and 18 521 respectively) across all provinces where the district was ‘unknown’ because the place of death was documented as the hospital name rather than the place name, and Stats SA do not currently code hospital names according to districts. Since these seem to be related to cases dying in hospital (rather than random deaths), there was concern that this would introduce bias into the relative distribution of deaths and YLLs by cause. It does result in reduced mortality rates in 2011 and 2012 (including age-standardised mortality rates) since a substantial number of deaths cannot be attributed to districts. Among the metros, this was clearly apparent for Nelson Mandela Bay and Buffalo City (EC), although it was more difficult to assess the impact in the other metros. In addition, the number of deaths recorded in eThekwini (KZN) declined by 38.8% between 2012 and 2013, with a distinct fall off over the year that is quite clearly different from the usual seasonal trends (Figure 1). For this reason mortality rates for the metros were only reported until 2012.

Figure 1: Monthly trends in deaths, eThekwini 2008 until 2013

p Bradshaw D, Dorrington RE, Laubscher R. Rapid Mortality Surveillance Report 2011. Cape Town: Medical Research Council, 2012. www.mrc.ac.za/bod/RapidMortality2011.pdf [accessed 30 Oct 2012].

q Maletela Ntoane-Nkhazi, Statistics SA, personal communication [25 Oct 2012].

221


Table 2: Number of registered deaths and stillbirths, 2008 until 2013

Year Deaths Stillbirths TotalUnknown

districtN %

2008 597 016 14 930 611 946 4 061 0.72009 582 024 14 274 596 298 4 480 0.82010 549 925 14 989 564 914 6 704 1.22011 514 486 14 212 528 698 23 536 4.62012 491 100 14 663 505 763 18 521 3.82013 458 933 14 604 473 537 9 620 2.1Total 3 193 484 87 672 3 281 156 66 922 2.0

Data quality

The two main indicators of data quality include completeness of registration (which is unknown at district level), and the percentage of deaths classified to ill-defined causes and ‘garbage codes’ as described earlier. The annual fluctuations in numbers of deaths by district and the changing proportions of deaths from ’unknown district’ by year suggest that completeness at district level is variable and that trends need to be interpreted with caution. Internationally, the recommended standard is less than 10% classification to ill-defined causes and garbage codes.r In South Africa, the total percentage of deaths coded to ill-defined causes and garbage codes declined from 39% in 1998s to 29.0% in 2008, and then +declined again slightly to 28.5% in 2013. However, there have been marked improvements in both ill-defined causes (from 9.5% to 7.1%) and garbage codes (from 16.1% to 12.5%) in the Western Cape, in contrast to other provinces over this period (Figure 2). This may reflect the increased efforts to improve death certification in the Western Cape (WC) as a result of the implementation of a local mortality surveillance system in that province. In order to reduce the proportion of ill-defined deaths it will be imperative to train doctors in death certification on the one hand, and to consider implementing alternative methods of establishing the underlying cause of death in rural areas, such as the use of verbal autopsy, on the other.

For the purposes of this report, the proportion of deaths coded to ill-defined causes was used as an indicator of the quality of mortality data. Ill-defined causes were reported for 14.0% of deaths in South Africa, and ranged from 4.5% (Eden, WC) to 52.4% (Alfred Nzo, EC) across districts (see Figure 3 and Map 1). In 2013, the percentage of ill-defined deaths in the eight metros was 13.6%. As might be expected, the percentage of ill-defined deaths was lowest in the districts within the highest socio-economic quintiles (SEQs 4-5)t and highest in districts in the lowest SEQs, probably reflecting lack of access to medical doctors to certify the cause of death. In such situations traditional headmen are allowed to certify natural deaths and reporting of ill-defined causes is more likely.

Garbage codes were reported for 14.5% of deaths in South Africa and ranged from 7.2% (John Taolo Gaetsewe, NC) to 20.1% (Nkangala, MP) (Figure 4). In Alfred Nzo (EC), OR Tambo (EC), Vhembe (LP) and Joe Gqabi (EC), more than 40% of deaths were coded to ill-defined causes and garbage codes. Interestingly, the percentage of garbage codes was highest in districts within the highest socio-economic quintiles (SEQs 4-5) and lowest in districts in the lowest SEQs (Figure 5). This may reflect better access to health services and medical information but poor certification practices on behalf of the doctors.

r Mathers CD, Inoue M, Ma Fat D, Rao C, et al. Counting the dead and what they died from: an assessment of the global status of cause of death data. Bull World Health Organ. 2004; 83:171-177.

s Pillay-van Wyk V, Bradshaw D, Groenewald P, Laubscher R. Improving the quality of medical certification of cause of death: the time is now! S Afr Med J. 2011 Sep 5;101(9):626.

t See Introduction and overview for details of the deprivation index and socio-economic quintiles [Page v].

Coding (group)

2008 2009 2010 2011 2012 2013Year of Year_date

IllDefined

Garbage

8%

10%

12%

14%

16%

18%

20%

22%

% o

f Tot

al D

eath

s

8%

10%

12%

14%

16%

18%

20%

22%

% o

f Tot

al D

eath

s

16.0% 16.8%

20.7%

13.0%

14.4%

7.1%

20.4%

9.4%9.5%9.5%

18.7%

11.4%

14.0%

15.1%14.6%

14.9%

12.9% 13.0%

12.6%

16.1%

12.1%

10.6%

14.8%

12.1%11.1%

13.5%

15.5%

14.0%

10.2%

16.6%

12.5%

17.7%

13.5%

Percentage of deaths ill-defined by provinceCoding (group)

IllDefined

Garbage

ProvEC

FS

GP

KZN

LP

MP

NC

NW

WC

222


Figure 2: Trend in ill-defined deaths and garbage codes by province 2008 until 2013

Percentage of deaths ill−defined by district, 2013

Percentage [Source: Stats SA Causes of Death]

A Nzo: DC44Vhembe: DC34

OR Tambo: DC15Joe Gqabi: DC14

JT Gaetsewe: DC45Johannesburg: JHB

Mangaung: MANEkurhuleni: EKU

Xhariep: DC16NM Molema: DC38

Mopani: DC33eThekwini: ETH

Harry Gwala: DC43Amathole: DC12Bojanala: DC37

West Rand: DC48Lejweleputswa: DC18

Capricorn: DC35S Baartman: DC10

uMzinyathi: DC24uMkhanyakude: DC27

Zululand: DC26C Hani: DC13

RS Mompati: DC39uMgungundlovu: DC22

Dr K Kaunda: DC40Pixley ka Seme: DC7

Waterberg: DC36Frances Baard: DC9

G Sibande: DC30Ugu: DC21

Ehlanzeni: DC32Namakwa: DC6

Nkangala: DC31iLembe: DC29

Buffalo City: BUFuThungulu: DC28

T Mofutsanyana: DC19Cape Winelands: DC2

Sedibeng: DC42ZF Mgcawu: DC8West Coast: DC1

N Mandela Bay: NMAOverberg: DC3

Cape Town: CPTSekhukhune: DC47

Tshwane: TSHFezile Dabi: DC20

uThukela: DC23Amajuba: DC25

Central Karoo: DC5Eden: DC4

10 20 30 40 50

NHI

NHI

NHI

NHI

NHI

NHINHI

NHI

NHINHI

NHI

16.915.4

8.4

8.5

15.3

13.3

5.4

12.7

13.3

6.3

9.7

11.8

20.4

7.1

18.5

14.7

7.7

14.7

8.2

6.9

18.5

7.4

18.2

52.4

13.6

14.9

12.0

29.731.7

8.4

33.8

13.9

10.4

7.2

9.7

9.0

9.6 9.6

10.4

7.7

10.0

28.4

14.9

18.0

11.8

11.1

7.3

7.6

8.1

7.4

4.5 4.9 SA avg: 14

ProvincesECFSGPKZNLPMPNCNWWC

223


Figure 3: Percentage of deaths ill-defined by district, 2013

224


Map 1: Percentage of deaths ill-defined by district, 2013

Percentage of deaths garbage codes by district, 2013

Percentage [Source: Stats SA Causes of Death]

Nkangala: DC31Tshwane: TSHiLembe: DC29

West Rand: DC48Sedibeng: DC42eThekwini: ETH

Johannesburg: JHBBuffalo City: BUF

Fezile Dabi: DC20Ekurhuleni: EKUuThukela: DC23

Dr K Kaunda: DC40Cape Winelands: DC2

uThungulu: DC28Mangaung: MAN

Xhariep: DC16uMgungundlovu: DC22

G Sibande: DC30T Mofutsanyana: DC19

Bojanala: DC37Amathole: DC12

Ugu: DC21OR Tambo: DC15

Amajuba: DC25West Coast: DC1

Central Karoo: DC5RS Mompati: DC39

C Hani: DC13NM Molema: DC38

Mopani: DC33Namakwa: DC6

S Baartman: DC10uMzinyathi: DC24Ehlanzeni: DC32Joe Gqabi: DC14

Zululand: DC26Waterberg: DC36

Lejweleputswa: DC18N Mandela Bay: NMA

Cape Town: CPTPixley ka Seme: DC7

uMkhanyakude: DC27Harry Gwala: DC43

Frances Baard: DC9Eden: DC4

Overberg: DC3Capricorn: DC35

Sekhukhune: DC47ZF Mgcawu: DC8

Vhembe: DC34A Nzo: DC44

JT Gaetsewe: DC45

5 10 15 20

NHI

NHI

NHI

NHI

NHI

NHI

NHI

NHI

NHI

NHI

NHI

14.0

13.6

13.0

13.9

8.9

12.3

16.8

13.2

14.4

12.5

14.2

16.6

14.5

18.518.8

16.3

17.0

19.6

13.9

14.2

15.7

13.2

13.9

12.9

11.9

14.5

19.0

11.7

17.5

13.3

9.3

10.7

12.8

10.5

14.2

20.1

13.2

13.2

12.1

9.7

11.4

7.2

14.0

13.4

13.8

14.8

12.1

13.9

14.6

10.811.2

13.9

SA avg: 14.5

ProvincesECFSGPKZNLPMPNCNWWC

225


Figure 4: Percentage of deaths with garbage codes by district, 2013

IndicatorShort Year_date (Calendar Year)

2008 2009 2010 2011 2012 2013

Cat

egor

y

13_B

urde

n of

dis

ease

Garbagecodes

Ill-defineddeaths

11

12

13

14

15

16In

dica

tor v

alue

10

12

14

16

18

20

Indi

cato

r val

ue

12.8

13.6

15.0

12.7

16.0

12.9

11.8

15.9

10.8

15.9

19.4

11.8

20.7

10.9

9.9

15.5

13.1

16.8

10.3

14.0

Indicator value by SEQ (weighted average of data by district quintile)SEQ

SEQ 1 (most deprived)

SEQ 2 (deprived)

SEQ 3

SEQ 4 (well off)

SEQ 5 (least deprived)

S E QS E Q 1 (most deprived)

S E Q 2 (deprived)

S E Q 3

S E Q 4 (well off)

S E Q 5 (least deprived)

226


Figure 5: Garbage codes and ill-defined deaths by SEQ (weighted average of data by district quintile)

Death Year Rank_YLL Nbdcodename

0% 2% 4% 6% 8% 10% 12% 14% 16% 18%% of Total YLL

2008 1 Tuberculosis

2 Lower respiratory infections

3 HIV/AIDS

4 Diarrhoeal diseases

5 Cerebrovascular disease

6 Ischaemic heart disease

7 Hypertensive heart disease

8 Meningitis/encephalitis

9 Accidental gunshot

10 Road injuries

2013 1 HIV/AIDS

2 Tuberculosis






8 Diabetes mellitus

9 Road injuries

10 Accidental threats to breathing

17.2%

12.1%

11.7%

10.7%

3.7%

2.8%

2.4%

2.3%

2.2%

2.1%

15.5%

12.4%

8.3%

5.7%

4.6%

3.3%

3.3%

2.8%

2.6%

2.4%

Leading YLLs (single causes), South Africa

227


Leading causes of premature mortality

The results presented here differ from the results presented in the Stats SA 2008 to 2013 cause of death reportsb,c,d,e,f,g in that ill-defined causes have been redistributed across other specified causes, and specific causes of injury are presented. It is important to note that a large proportion of HIV deaths have been misattributed to immediate causes of death such as TB, diarrhoeal diseases and lower respiratory infections,u,v and that since many injury-related deaths are misclassified to ill-defined intent,w the ranking of injury-related causes may be unreliable. In 2008 and 2013, the four leading single causes of YLLs in South Africa were HIV-related conditions, TB, pneumonia, and diarrhoea, suggesting that HIV-related mortality remains the leading cause of YLLs in the majority of districts in South Africa (Figure 6). Also in the top 10 leading causes of YLLs across South Africa are cerebrovascular diseases, hypertensive heart disease, ischaemic heart disease, diabetes and road injuries. With some minor differences in ranking between provinces, preterm birth complications (KZN, North West (NW) and NC), chronic obstructive pulmonary disease (COPD) (WC and NC), lung cancer (WC), and meningitis and encephalitis (LP and MP), are also among the top 10 causes of premature mortality across most districts in South Africa (Figure 7 and Figure 8).

The extent of the problem with regard to ill-defined injuries is illustrated by comparing the leading causes from the Western Cape mortality surveillance system,x where injury data are collected directly from the mortuaries, with the Stats SA injury data that reflect what is reported on the death certificate, and that often do not report the manner of death or the intent. This shows that the ranking and proportion of specific causes of injury related deaths based on the Stats SA data are different from the Western Cape mortality surveillance data. The Western Cape surveillance data should be more accurate with regard to the causes of injury related deaths (Table 3). In the Western Cape, premature mortality due to interpersonal violence is ranked 2nd according to the Western Cape mortality surveillance, accounting for 9.2% of YLLs. Using Stats SA data, it ranks 5th, accounting for only 4.8% of YLLs. Road injury deaths rank 6th in the Western Cape surveillance data (4.6% of YLLs), while this cause does not appear in the top 10 in the Stats SA data. Deaths due to accidental gunshots rank 8th in the Stats SA data (3.4% of YLLs). According to information from the forensic mortuaries in the Western Cape, these deaths are much more likely to be due to interpersonal violence than accidents. There was only one death due to accidental gunshot in 2010, four in 2011 and zero in 2012.

Figure 6: Leading causes of Years of Life Lost (YLLs) for South Africa, 2008 and 2013

u Groenewald P, Nannan N, Bourne D, Laubscher R, Bradshaw D. Identifying deaths from AIDS in South Africa. AIDS. 2005; 19:193-201.

v Yudkin PL, Burger EH, Bradshaw D, Groenewald P, Ward AM, Volmink J. Deaths caused by HIV disease under-reported in South Africa. AIDS. 2009 Jul 31; 23(12):1600-2.

w Norman R, Matzopoulos R, Groenewald P, Bradshaw D. The high burden of injuries in South Africa. Bull World Health Organ. 2007;85:695-702.

x Groenewald P, Msemburi W, Morden E, Zinyakatira N, Neethling I, Daniels J, et al. Western Cape mortality profile 2011. Cape Town: South African Medical Research Council, 2014.

2013

0% 5% 10% 15% 20%% of Total YLLs

EC 1 HIV/AIDS

2 Tuberculosis




6 Interpersonal violence


8 Diabetes mellitus

9 Road injuries


FS 1 HIV/AIDS

2 Tuberculosis





7 Road injuries



10 Nephritis/nephrosis

GP 1 HIV/AIDS

2 Tuberculosis








10 Diabetes mellitus

KZN 1 HIV/AIDS

2 Tuberculosis




6 Diabetes mellitus





LP 1 HIV/AIDS


3 Tuberculosis



6 Road injuries

7 Diabetes mellitus


9 Meningitis/encephalitis

10 Nephritis/nephrosis

17.7%

14.9%

6.0%

4.9%

4.6%

3.6%

3.4%

2.8%

2.7%

2.6%

14.4%

11.7%

10.8%

6.3%

4.9%

3.9%

3.3%

3.1%

2.8%

2.5%

13.4%

10.8%

9.3%

4.1%

4.0%

4.0%

3.5%

2.9%

2.9%

2.6%

18.5%

15.7%

6.4%

6.0%

5.2%

3.0%

2.9%

2.7%

2.5%

2.4%

13.7%

11.8%

10.9%

10.3%

4.2%

4.2%

4.0%

3.6%

3.1%

2.9%

Leading YLLs (single causes) by province2013

0% 10% 20%% of Total YLLs

MP 1 HIV/AIDS

2 Tuberculosis




6 Road injuries


8 Diabetes mellitus


10 Septicaemia

NC 1 HIV/AIDS

2 Tuberculosis




6 Road injuries




10 COPD

NW 1 HIV/AIDS

2 Tuberculosis





7 Diabetes mellitus


9 Road injuries


WC 1 HIV/AIDS

2 Tuberculosis




6 Trachea/bronchi/lung


8 COPD


10 Diabetes mellitus

18.0%

13.4%

10.4%

7.3%

4.3%

3.8%

3.5%

3.0%

2.6%

1.9%

19.6%

10.6%

6.8%

4.8%

4.3%

4.0%

4.0%

3.9%

3.2%

2.7%

16.1%

12.0%

11.4%

6.7%

5.4%

4.1%

2.6%

2.5%

2.2%

2.1%

11.5%

8.7%

6.8%

5.3%

5.0%

4.3%

4.1%

3.6%

3.2%

2.7%

Leading YLLs (single causes) by province

228


Figure 7: 10 leading YLLs by province, 2013

Nbdcodename

EC

A N

zo: D

C44

Am

atho

le: D

C12

Buf

falo

City

: BU

F

C H

ani:

DC

13

Joe

Gqa

bi: D

C14

N M

ande

la B

ay: N

MA

OR

Tam

bo: D

C15

S B

aartm

an: D

C10

FS

Fezi

le D

abi:

DC

20

Lejw

elep

utsw

a: D

C18

Man

gaun

g: M

AN

T M

ofut

sany

ana:

DC

19

Xha

riep:

DC

16

GP

Eku

rhul

eni:

EK

U

Joha

nnes

burg

: JH

B

Sed

iben

g: D

C42

Tshw

ane:

TS

H

Wes

t Ran

d: D

C48

KZN

Am

ajub

a: D

C25

Har

ry G

wal

a: D

C43

Ugu

: DC

21

Zulu

land

: DC

26

eThe

kwin

i: E

TH

iLem

be: D

C29

uMgu

ngun

dlov

u: D

C22

uMkh

anya

kude

: DC

27

uMzi

nyat

hi: D

C24

uThu

kela

: DC

23

uThu

ngul

u: D

C28

LP

Cap

ricor

n: D

C35

Mop

ani:

DC

33

Sek

hukh

une:

DC

47

Vhe

mbe

: DC

34

Wat

erbe

rg: D

C36

MP

Ehl

anze

ni: D

C32

G S

iban

de: D

C30

Nka

ngal

a: D

C31

NC

Fran

ces

Baa

rd: D

C9

JT G

aets

ewe:

DC

45

Nam

akw

a: D

C6

Pix

ley

ka S

eme:

DC

7

ZF M

gcaw

u: D

C8

NW

Boj

anal

a: D

C37

Dr K

Kau

nda:

DC

40

NM

Mol

ema:

DC

38

RS

Mom

pati:

DC

39

WC

Cap

e To

wn:

CP

T

Cap

e W

inel

ands

: DC

2

Cen

tral K

aroo

: DC

5

Ede

n: D

C4

Ove

rber

g: D

C3

Wes

t Coa

st: D

C1

HIV/AIDS

Tuberculosis

Lower respiratory infections

Cerebrovascular disease

Diarrhoeal diseases

Hypertensive heart disease

Ischaemic heart disease

Diabetes mellitus

Accidental threats to breathing

Nephritis/nephrosis

Interpersonal violence

Road injuries

Preterm birth complications

COPD

Endocrine nutritional,blood, immu..

Meningitis/encephalitis

Accidental gunshot

Septicaemia

Epilepsy

Asthma

16

18

7

12

6

14

10

9

5

8

11

3

4

2

1

17

16

19

10

14

18

6

7

12

8

11

9

3

5

4

2

1

15

18

20

11

10

14

19

8

9

12

6

4

5

16

3

7

2

1

14

20

12

15

19

7

10

11

13

8

9

4

6

5

3

2

1

8

12

17

16

14

20

13

6

18

10

9

11

7

5

4

3

2

1

13

20

19

7

9

4

14

10

6

8

12

11

3

5

2

1

8

13

17

18

12

11

7

20

9

10

16

6

4

5

3

2

1

18

17

11

20

16

9

6

12

10

7

14

8

4

5

3

2

1

18

12

17

14

16

9

8

11

10

13

6

7

5

4

3

2

1

17

15

18

12

7

11

14

10

8

9

6

4

5

2

3

1

12

19

17

14

16

11

6

7

13

10

8

9

5

4

3

2

1

15

14

13

16

12

7

6

10

19

11

8

9

4

5

1

3

2

20

19

15

11

16

12

7

14

10

13

9

8

5

4

6

2

1

3

18

7

15

13

17

16

10

11

14

8

12

4

9

5

6

2

3

1

13

12

14

17

16

9

10

11

8

4

5

7

6

3

2

1

19

17

11

13

18

16

15

10

12

8

9

7

6

4

5

1

2

3

10

4

16

19

13

12

7

9

11

6

15

8

5

3

2

1

17

7

11

16

14

12

15

13

8

9

6

10

4

5

3

2

1

20

19

13

12

15

8

6

11

10

16

9

14

7

3

5

4

2

1

18

16

17

7

14

15

10

11

12

13

9

6

8

3

5

4

2

1

17

10

14

19

7

6

13

12

20

9

11

8

4

5

3

2

1

18

13

16

17

12

6

9

15

10

8

11

7

3

4

5

2

1

19

14

12

13

15

16

10

11

8

6

7

9

4

3

5

2

1

19

16

6

14

13

7

15

12

11

9

8

10

3

4

5

1

2

18

14

6

13

16

11

19

15

8

10

9

5

12

7

4

3

1

2

15

18

11

16

12

9

10

13

6

8

14

7

3

5

4

1

2

13

17

15

14

20

16

12

8

11

6

7

10

9

5

3

4

1

2

13

14

12

16

11

8

10

7

18

6

17

9

3

5

4

2

1

20

17

14

15

10

7

13

9

12

8

11

6

4

5

3

2

1

17

10

16

15

11

5

12

13

8

9

6

4

7

3

2

1

12

13

15

17

9

7

8

5

19

11

3

6

4

2

1

17

20

9

10

15

14

18

5

11

12

8

16

7

3

6

2

4

1

18

5

17

10

11

16

6

12

7

14

9

4

8

2

1

3

16

11

14

18

13

5

9

10

7

12

6

4

8

2

3

1

20

14

19

17

16

18

10

12

11

8

7

5

4

6

2

3

1

20

18

14

16

17

13

5

11

12

9

8

10

7

4

6

3

2

1

7

12

10

14

15

6

18

13

17

8

11

9

4

5

3

2

1

17

9

11

6

5

20

16

12

8

10

4

7

3

2

1

15

8

13

9

5

18

17

12

6

10

7

4

3

2

1

18

20

4

9

6

17

8

13

3

11

5

7

2

1

14

13

20

7

5

9

19

12

10

6

4

8

3

2

1

14

13

18

7

9

12

16

10

6

8

5

4

3

2

1

20

19

17

10

18

7

13

9

14

8

5

4

6

2

3

1

16

9

14

13

10

7

19

11

20

8

15

5

4

6

1

3

2

15

16

20

13

12

19

11

6

9

8

7

4

5

3

2

1

13

15

20

19

14

16

8

11

12

9

7

10

5

4

6

3

2

1

13

6

15

5

14

9

12

3

10

16

4

7

1

2

7

19

5

8

11

20

12

2

14

18

4

9

3

1

17

7

16

8

10

11

14

9

3

12

13

4

6

2

1

17

5

12

3

6

13

16

15

4

8

11

7

10

2

1

13

5

7

11

8

10

3

16

15

4

9

2

1

20

4

9

15

12

5

11

13

10

3

14

17

6

8

2

1

Leading YLLs, rank by district, 2013

Broad causeComm_mat_peri_nut

HIV and TB

Injury

NCD

229


Figure 8: Ranking of 20 leading YLLs by district, 2013

Table 3: Ranking of injury-related causes of premature death – a comparison between Western Cape Surveillance data and Western Cape Stats SA data, 2012

Rank Western Cape Surveillance Western Cape Stats SA1 HIV and AIDS 12.0% HIV and AIDS 12.2%2 Interpersonal violence 9.2% Tuberculosis 9.7%3 Tuberculosis 7.4% Ischaemic heart disease 6.9%4 Ischaemic heart disease 6.5% Cerebrovascular disease 5.7%5 Cerebrovascular disease 5.2% Interpersonal violence 4.8%6 Road injuries 4.6% Tracheal/bronchial/lung conditions 3.7%7 Diabetes mellitus 4.3% COPD 3.5%8 COPD 3.8% Accidental gunshot 3.4%9 Tracheal/bronchial/lung conditions 3.7% Lower respiratory infections 3.3%

10 Lower respiratory infections 3.5% Diabetes mellitus 2.7%

Trends in leading causes of premature mortality

Tuberculosis remained the leading cause of premature mortality from 2008 to 2011 (Figure 9). HIV and AIDS ranked 3rd in 2008 after lower respiratory infections, but moved to 2nd position in 2009, where it remained until it moved to 1st place in 2012. Lower respiratory infections ranked 3rd in 2013. Meningitis/encephalitis dropped out of the top 10 in 2011. Diabetes mellitus moved from 10th to 8th place. Road injuries moved up from 10th to 9th position, and hypertensive heart disease moved from 7th position in 2008 to 6th place in 2013.

In the Eastern Cape (Figure 10), HIV and AIDS moved up to become the leading cause of premature mortality in all districts from 2008 to 2013. Cerebrovascular disease rose in the ranks in all districts, and diabetes moved up in all Eastern Cape districts with the exception of Amathole.

In the Free State, HIV and AIDS rose in the ranking in all districts between 2008 and 2013. Diabetes increased in Fezile Dabi and Mofutsanyana, and cerebrovascular disease increased in the ranking in Fezile Dabi, Mangaung and Xhariep.

In Gauteng, HIV and AIDS went up in the ranking in all districts between 2008 and 2013. Ischaemic heart disease increased in West Rand, Tshwane and Ekurhuleni. Cerebrovascular disease moved up in the ranking in Ekurhuleni, Johannesburg and Tshwane, and dropped in West Rand.

In KwaZulu-Natal, HIV and AIDS moved up in the YLL ranking in Amajuba, Harry Gwala, Umzinyathi, Uthukela, Uthungulu, Zululand and uMgungundlovu. Cerebrovascular disease moved up in the ranking in eThekwini, iLembe, Ugu, Zululand, uMgungundlovu and uMkhanyakude districts, and diabetes increased in Amajuba, Uthukela, Harry Gwala, iLembe, Uthungulu, Ugu, Umgungundlovu, Umkhanyakude, Uthukela, and Zululand. In Umzinyathi, preterm birth complications dropped from 6th to 7th in the ranking.

2008 2009 2010 2011 2012 2013 2014Year of death

Interpersonal violence10 (2.1%)

Hypertensive heart disease7 (2.4%)


Accidental strangulation10 (2.4%)

Accidental strangulation10 (2.4%)

Accidental gunshot9 (2.2%)

Meningitis/encephalitis8 (2.3%)


Road injuries10 (2.1%)


Ischaemic heart disease6 (2.8%)


Diabetes mellitus10 (2.1%)


Cerebrovascular disease5 (3.7%)


Diarrhoeal diseases4 (10.7%)


Lower respiratory infections2 (12.1%)


HIV/AIDS3 (11.7%)

HIV/AIDS1 (15.5%)

Tuberculosis1 (17.2%)


Trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs2.1%

5.0%

10.0%

17.2%

230


In Limpopo, diarrhoeal disease went down in the ranking in Capricorn, Mopani, Sekhukhune, Vhembe and Waterberg, while HIV and AIDS moved up in all districts. Tuberculosis moved up in Mopani, and diabetes mellitus moved up in the ranking in Capricorn, Mopani, Sekhukhune, Vhembe and Waterberg.

In Mpumalanga, HIV and AIDS moved up in the ranking in Ehlanzeni, Gert Sibande and Nkangala, while diarrhoea went down in all districts. Cerebrovascular disease went up in Nkangala, and diabetes went up in all districts.

In North West, HIV and AIDS moved up in the ranking in all districts between 2008 and 2013. Diabetes moved up in the ranking in Bojanala, Dr K Kaunda and NM Molema. Preterm birth complications dropped in the ranking in NM Molema but increased in RS Mompati.

In the Northern Cape, HIV and AIDS moved up in the ranking in all districts between 2008 and 2013. Diarrhoeal disease dropped from 3rd to 7th place in Pixley ka Seme, and cerebrovascular disease went up in the ranking in Frances Baard, Namakwa and Pixley ka Seme.

HIV and AIDS increased in the ranking in all Western Cape districts between 2008 and 2013. Cerebrovascular disease decreased in Cape Town, Central Karoo and West Coast. Chronic obstructive pulmonary disease increased in the ranking in Cape Town Metro, Overberg and West Coast. Lung cancer increased in the ranking in all Western Cape districts except West Coast. Diarrhoeal disease dropped in the ranking in all districts, while lower respiratory infections increased in the Cape Winelands, Eden and Overberg.

Figure 9: Trends in 10 leading YLLs, South Africa, 2008–2013

District

2008 2009 2010 2011 2012 2013 2014Year of death

A Nzo: DC44

Amathole:DC12

Buffalo City:BUF

C Hani: DC13

Joe Gqabi:DC14



HIV/AIDS1 (19.5%)


HIV/AIDS2 (13.4%)










Other respiratory6 (2.8%)







Accidental threats to breathing10 (2.0%)















HIV/AIDS1 (15.5%)


HIV/AIDS2 (15.6%)



COPD10 (2.2%)

Asthma8 (3.5%)

Asthma7 (3.6%)



Nephritis/nephrosis9 (2.7%)




COPD8 (2.4%)


COPD7 (3.1%)



HIV/AIDS1 (15.0%)







HIV/AIDS2 (8.0%)





Cardiomyopathy10 (2.4%)





HIV/AIDS1 (18.2%)







HIV/AIDS3 (11.2%)







Asthma8 (3.0%)

Asthma9 (2.4%)




EC trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.7%

5.0%

10.0%

15.0%

20.0%

22.7%

231


Figure 10: Trends in 10 leading YLLs by district, 2008-2013

District

2008 2009 2010 2011 2012 2013 2014Year of death

Fezile Dabi:DC20

Lejweleputswa: DC18

Mangaung:MAN

TMofutsanyana:DC19

Xhariep: DC16



HIV/AIDS4 (9.9%)




HIV/AIDS3 (9.9%)














Endocrine nutritional,blood, immune7 (2.9%)

Preterm birth complications10 (2.6%)






HIV/AIDS2 (12.2%)








HIV/AIDS4 (9.7%)









HIV/AIDS1 (15.0%)


HIV/AIDS3 (12.8%)




















Septicaemia10 (2.0%)













HIV/AIDS3 (13.5%)

HIV/AIDS1 (18.1%)










FS trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.7%

5.0%

10.0%

15.0%

20.0%

23.4%

232


District

2008 2009 2010 2011 2012 2013 2014Year of death

TMofutsanyana:DC19

Xhariep: DC16


HIV/AIDS4 (8.0%)

HIV/AIDS1 (13.3%)






















Protein-energy malnutrition9 (1.9%)

FS trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.7%

5.0%

10.0%

15.0%

20.0%

23.4%

233


District

2008 2009 2010 2011 2012 2013 2014Year of death

Ekurhuleni:EKU

Johannesburg:JHB

Sedibeng:DC42

Tshwane: TSH

West Rand:DC48














HIV/AIDS1 (15.3%)





HIV/AIDS3 (12.1%)












HIV/AIDS1 (13.0%)

HIV/AIDS1 (13.2%)








Septicaemia10 (2.4%)



Septicaemia9 (2.5%)









HIV/AIDS3 (10.6%)

HIV/AIDS4 (6.4%)


















HIV/AIDS3 (9.7%)



HIV/AIDS1 (13.1%)

















GP trends in leading YLLs, rank (% of total YLLs)% YLLs

1.9%

5.0%

10.0%

15.0%

19.3%


HIV and TB

Injury

NCD

234


District

2008 2009 2010 2011 2012 2013 2014Year of death

Tshwane: TSH

West Rand:DC48














HIV/AIDS4 (11.0%)

HIV/AIDS1 (12.5%)








GP trends in leading YLLs, rank (% of total YLLs)% YLLs

1.9%

5.0%

10.0%

15.0%

19.3%


HIV and TB

Injury

NCD

235


District

2008 2009 2010 2011 2012 2013 2014Year of death

Amajuba: DC25

eThekwini: ETH

Harry Gwala:DC43

iLembe: DC29

Ugu: DC21

HIV/AIDS1 (20.7%)


HIV/AIDS4 (8.2%)








































HIV/AIDS2 (13.5%)

HIV/AIDS2 (11.3%)









HIV/AIDS1 (19.4%)

HIV/AIDS3 (12.9%)










Asthma9 (2.4%)

Asthma6 (4.0%)






















HIV/AIDS2 (17.6%)

HIV/AIDS2 (18.8%)



KZN trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.4%

10.0%

20.0%

31.1%

236


District

2008 2009 2010 2011 2012 2013 2014Year of death

iLembe: DC29

Ugu: DC21

uMgungundlovu: DC22

uMkhanyakude:DC27

uMzinyathi:DC24

uThukela:DC23













HIV/AIDS2 (17.7%)

HIV/AIDS2 (15.6%)







Asthma8 (2.1%)

Asthma9 (2.2%)



















HIV/AIDS1 (16.9%)

HIV/AIDS2 (11.2%)



























HIV/AIDS1 (29.0%)

HIV/AIDS1 (31.1%)

















HIV/AIDS2 (12.9%)




HIV/AIDS1 (19.3%)







HIV and TB

Injury

NCD

% YLLs1.4%

10.0%

20.0%

31.1%

237


District

2008 2009 2010 2011 2012 2013 2014Year of death

uMzinyathi:DC24

uThukela:DC23

uThungulu:DC28

Zululand: DC26

HIV/AIDS4 (11.7%)









HIV/AIDS1 (19.1%)





























HIV/AIDS1 (21.9%)

HIV/AIDS2 (18.8%)













HIV/AIDS3 (10.4%)



HIV/AIDS2 (19.7%)

















HIV and TB

Injury

NCD

% YLLs1.4%

10.0%

20.0%

31.1%

238


District

2008 2009 2010 2011 2012 2013 2014Year of death

Capricorn:DC35

Mopani: DC33

Sekhukhune:DC47

Vhembe:DC34

Waterberg:DC36

HIV/AIDS1 (14.1%)

HIV/AIDS4 (8.4%)






















Nephritis/nephrosis9 (2.1%) Hypertensive heart disease

9 (2.5%)






HIV/AIDS3 (10.2%)



HIV/AIDS4 (9.2%)








Septicaemia9 (1.7%)




HIV/AIDS1 (16.0%)









HIV/AIDS4 (8.6%)












Septicaemia9 (2.7%)

Septicaemia9 (2.2%)

HIV/AIDS1 (13.2%)




HIV/AIDS4 (5.2%)

















LP trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.4%

5.0%

10.0%

15.0%

20.0%

26.7%

239


District

2008 2009 2010 2011 2012 2013 2014Year of death

Vhembe:DC34

Waterberg:DC36




HIV/AIDS4 (11.5%)















HIV/AIDS1 (15.9%)



LP trends in leading YLLs, rank (% of total YLLs)


HIV and TB

Injury

NCD

% YLLs1.4%

5.0%

10.0%

15.0%

20.0%

26.7%

240


District

2008 2009 2010 2011 2012 2013 2014Year of death

Ehlanzeni:DC32

G Sibande:DC30

Nkangala:DC31




Septicaemia8 (1.9%)

HIV/AIDS4 (12.5%)







HIV/AIDS1 (18.5%)


Septicaemia7 (3.3%)









HIV/AIDS4 (12.3%)






HIV/AIDS1 (20.4%)
























HIV/AIDS4 (9.5%)

HIV/AIDS1 (15.0%)











COPD9 (2.4%)

Poisonings (including herbal)9 (3.0%)


MP trends in leading YLLs, rank (% of total YLLs)% YLLs

1.4%

5.0%

10.0%

15.0%

20.4%


HIV and TB

Injury

NCD

241


District

2008 2009 2010 2011 2012 2013 2014Year of death

Ehlanzeni:DC32

G Sibande:DC30

Nkangala:DC31




Septicaemia8 (1.9%)

HIV/AIDS4 (12.5%)







HIV/AIDS1 (18.5%)


Septicaemia7 (3.3%)









HIV/AIDS4 (12.3%)






HIV/AIDS1 (20.4%)
























HIV/AIDS4 (9.5%)

HIV/AIDS1 (15.0%)











COPD9 (2.4%)

Poisonings (including herbal)9 (3.0%)


MP trends in leading YLLs, rank (% of total YLLs)% YLLs

1.4%

5.0%

10.0%

15.0%

20.4%


HIV and TB

Injury

NCD

242


District

2008 2009 2010 2011 2012 2013 2014Year of death

FrancesBaard: DC9

JT Gaetsewe:DC45

Namakwa:DC6

Pixley kaSeme: DC7

ZF Mgcawu:DC8



HIV/AIDS2 (12.9%)


HIV/AIDS1 (18.3%)
























HIV/AIDS4 (3.5%)

HIV/AIDS1 (29.7%)
















Birth asphyxia9 (2.1%)






HIV/AIDS2 (8.8%)





HIV/AIDS1 (9.6%)


COPD5 (4.7%)


COPD4 (5.8%)







Trachea/bronchi/lung10 (3.3%)





Cirrhosis of liver9 (2.4%)















COPD8 (3.7%)

COPD8 (3.3%)



HIV/AIDS1 (17.2%)

HIV/AIDS2 (11.5%)



NC trends in leading YLLs, rank (% of total YLLs)% YLLs

1.8%

10.0%

20.0%

29.7%


HIV and TB

Injury

NCD

243


District

2008 2009 2010 2011 2012 2013 2014Year of death

Pixley kaSeme: DC7

ZF Mgcawu:DC8 Road injuries

6 (4.5%)


COPD8 (3.7%)







HIV/AIDS2 (12.8%)



HIV/AIDS1 (17.7%)


COPD9 (3.2%)







Drowning9 (2.6%)

NC trends in leading YLLs, rank (% of total YLLs)% YLLs

1.8%

10.0%

20.0%

29.7%


HIV and TB

Injury

NCD

244


District

2008 2009 2010 2011 2012 2013 2014Year of death

Bojanala:DC37

Dr K Kaunda:DC40

NM Molema:DC38

RS Mompati:DC39


HIV/AIDS3 (12.7%)







HIV/AIDS1 (14.5%)

















HIV/AIDS1 (18.1%)


HIV/AIDS2 (12.8%)













Fires, hot substances10 (2.2%)




Fires, hot substances9 (2.0%)






HIV/AIDS2 (12.6%)








HIV/AIDS4 (6.7%)












Septicaemia9 (2.1%)

Septicaemia7 (2.9%)

HIV/AIDS4 (13.5%)


HIV/AIDS1 (22.7%)





















NW trends in leading YLLs, rank (% of total YLLs)% YLLs

1.6%

5.0%

10.0%

15.0%

20.0%

22.7%


HIV and TB

Injury

NCD

245


District

2008 2009 2010 2011 2012 2013 2014Year of death

Cape Town:CPT

CapeWinelands:DC2

Central Karoo:DC5

Eden: DC4

Overberg: DC3







HIV/AIDS1 (11.7%)


HIV/AIDS2 (11.6%)











COPD10 (2.5%)


COPD9 (2.8%)







COPD4 (5.4%)




COPD5 (5.7%)




HIV/AIDS1 (13.3%)


HIV/AIDS2 (8.9%)











HIV/AIDS3 (8.8%)

HIV/AIDS1 (12.4%)












COPD5 (7.2%)





COPD4 (5.7%)








HIV/AIDS1 (10.9%)


HIV/AIDS2 (7.3%)











COPD7 (3.9%)

COPD7 (3.9%)


WC trends in leading YLLs, rank (% of total YLLs)% YLLs

2.0%

5.0%

10.0%

15.0%

17.8%


HIV and TB

Injury

NCD

246


District

2008 2009 2010 2011 2012 2013 2014Year of death

Eden: DC4

Overberg: DC3

West Coast:DC1



Drowning10 (2.5%)


HIV/AIDS3 (6.3%)



COPD10 (3.7%)

Drowning4 (5.1%)





HIV/AIDS1 (10.1%)




COPD7 (4.8%)






HIV/AIDS4 (6.1%)





HIV/AIDS2 (8.7%)






COPD9 (3.4%)

Drowning10 (2.6%)

Drowning10 (2.8%)



COPD6 (4.8%)





WC trends in leading YLLs, rank (% of total YLLs)% YLLs

2.0%

5.0%

10.0%

15.0%

17.8%


HIV and TB

Injury

NCD

247


Death Year

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% 105%% of Total YLLs

200820092010201120122013

10.6%10.5%10.9%11.4%

12.5%13.1%

29.1%31.2%

32.0%34.5%

35.6%36.9%

29.0%29.1%

28.9%29.0%

29.0%27.9%

31.3%29.2%

28.2%25.2%

22.9%22.1%

Percentage of YLLs by broad cause, South Africa

BroadcauseInjuryNCDHIV and TBComm_mat_peri_nut

Prov

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%% of Total YLLs

WCGPECNCFS

KZNNWLPMP

Grand Total 36.9%27.9%22.1% 13.1%

12.9%14.2%13.9%

11.9%12.6%

10.3%10.2%

12.7%

51.1%40.4%37.1%36.4%

36.7%32.4%

34.4%32.7%

29.5%

24.2%32.6%

30.2%26.0%

34.2%28.1%

24.7%31.4%

20.2%10.7%22.5%

16.1%19.5%

25.3%20.8%

27.2%32.4%

26.4%

17.9%

Percentage of YLLs by broad cause, by province, 2013


248


Cause of death profile

South Africa still faces a quadruple burden of Comm/Mat/Peri/Nutr, HIV and TB, NCDs and injuries. However, the percentage of the burden due to HIV and TB and Comm/Mat/Peri/Nutr declined between 2008 and 2013 from 60% to 50%, with a corresponding increase in the burden due to NCDs (from 29% to 37%) and to a lesser extent injuries (from 11% to 13%) (see Figure 11).

Figure 11: Percentage of YLLs by broad cause, South Africa, 2008-2013

The quadruple burden varied across provinces, with the Western Cape having a higher proportion due to injury (16.8%) and NCDs (49.4%) than any other province (Figure 12). KwaZulu-Natal, Mpumalanga, Limpopo and North West had the highest proportions due to HIV and TB and Comm/Mat/Peri/Nutr (approximately 60%). Districts within the provinces reflect the provincial profiles. uMkanyakude (KZN) had the highest burden due to HIV and TB (43.5%), and Overberg (WC) the lowest (17.6%) (Figure 13). Districts falling into the highest socio-economic quintile (SEQ5) and metros had higher proportions of YLLs due to injuries and NCDs, while those falling into the lowest socio-economic quintiles had higher proportions of YLLs due to HIV and TB and Comm/Mat/Peri/Nutr (Map 2).

Figure 12: Percentage of YLLs by broad cause by province, 2013

District

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%% of Total YLLs

Overberg: DC3

Namakwa: DC6

Cape Town: CPT

Central Karoo: DC5

Cape Winelands: DC2

Eden: DC4

West Coast: DC1

Buffalo City: BUF

N Mandela Bay: NMA

Tshwane: TSH

Johannesburg: JHB

S Baartman: DC10

Pixley ka Seme: DC7

Amathole: DC12

Xhariep: DC16

uMgungundlovu: DC22

Mangaung: MAN

ZF Mgcawu: DC8

West Rand: DC48

eThekwini: ETH

Fezile Dabi: DC20

Sedibeng: DC42

C Hani: DC13

Frances Baard: DC9

Nkangala: DC31

Bojanala: DC37

Ekurhuleni: EKU

Dr K Kaunda: DC40

Joe Gqabi: DC14

OR Tambo: DC15

Capricorn: DC35

Lejweleputswa: DC18

Ugu: DC21

Waterberg: DC36

Vhembe: DC34

T Mofutsanyana: DC19

NM Molema: DC38

uThukela: DC23

Harry Gwala: DC43

Amajuba: DC25

uThungulu: DC28

iLembe: DC29

uMzinyathi: DC24

G Sibande: DC30

Sekhukhune: DC47

Mopani: DC33

Ehlanzeni: DC32

uMkhanyakude: DC27

A Nzo: DC44

RS Mompati: DC39

JT Gaetsewe: DC45

Zululand: DC26

10.5%

10.4%

15.2%

12.0%

15.7%

13.6%

19.3%

16.0%

10.7%

19.4%

11.4%

12.8%

10.6%

12.1%

14.0%

11.7%

13.0%

13.0%

10.2%

13.4%

12.4%

13.6%

10.4%

13.2%

12.1%

21.1%

11.9%

20.1%

15.6%

18.2%

16.3%

18.7%

13.6%

23.9%

13.6%

13.7%

10.4%

10.7%

11.5%

14.1%

12.8%

11.5%

12.9%

12.9%

12.3%

10.1%

11.6%

16.3%

15.5%

12.8%

16.9%

10.6%

27.8%

31.9%

38.8%

36.1%

44.6%

37.1%

50.6%

51.8%

35.3%

48.8%

36.2%

53.7%

29.0%

35.6%

38.5%

39.5%

37.4%

27.0%

32.0%

28.0%

35.1%

43.1%

24.6%

32.7%

41.1%

31.4%

47.0%

51.0%

32.7%

34.2%

29.9%

54.2%

40.8%

28.5%

42.0%

37.5%

29.5%

33.2%

46.0%

31.3%

40.7%

27.4%

27.5%

29.7%

29.4%

33.9%

32.8%

50.0%

37.3%

40.8%

36.2%

24.1%

37.1%

33.5%

29.7%

25.6%

28.6%

31.9%

19.1%

23.4%

23.8%

22.3%

32.0%

21.8%

34.5%

27.9%

29.2%

24.7%

29.9%

32.0%

35.8%

38.9%

32.9%

22.7%

38.4%

23.8%

26.5%

22.8%

29.3%

18.3%

26.3%

24.9%

37.4%

17.6%

28.4%

33.4%

32.5%

21.7%

25.6%

28.2%

22.9%

35.8%

29.2%

43.5%

32.2%

34.0%

35.4%

24.9%

27.8%

22.7%

23.3%

25.3%

28.6%

41.5%

24.6%

24.3%

16.4%

26.3%

11.1%

17.4%

11.0%

30.1%

20.4%

11.7%

25.9%

24.3%

18.3%

24.1%

19.7%

28.0%

21.9%

19.6%

19.6%

20.7%

26.6%

30.3%

20.2%

37.5%

11.8%

10.6%

25.4%

32.5%

16.3%

17.2%

29.8%

11.9%

27.1%

34.5%

27.9%

19.6%

18.8%

17.3%

17.6%

27.4%

23.4%

22.9%

31.1%

27.8%

11.0%

23.8%

8.3%

8.4%

8.3%

9.5%

9.6%

8.8%

Percentage of YLLs by broad cause, by district, 2013


YLLs sorted in ascending order of the combined proportion of Communicable and Maternal YLLs and YLLs due to HIV and TB.

249


Figure 13: Percentage of YLLs by broad causes, by district, 2013

250


Map 2: Percentage of YLLs by broad cause, by district, 2013Communicable diseases together with perinatal, maternal and nutritional conditions years of life lost

HIV and TB years of life lost

Non-communicable diseases years of life lost Injury years of life lost

Note: These percentages do not give any indication of the level of mortality due to these causes as would be provided by age-standardised mortality rates, but only of the relative proportion of all deaths in each district due to each broad group of causes. The percentage of YLLs for the four broad causes therefore totals 100% for each district.

Metro mortality rates

It is important to note the data challenges and inconsistencies pertaining to the district data as described at the beginning of the Results section when interpreting these results for the metros.

In 2012, comparative mortality ratios for all-cause mortality across the eight metros (with Nelson Mandela Bay Metro (EC) as the base), showed that Nelson Mandela Bay had the lowest all-cause mortality and Mangaung (FS) the highest, with 2.58 times the mortality experienced in Nelson Mandela Bay, after standardising for age (Figure 14). Mortality due to Comm/Mat/Per/Nutr showed the greatest variation between metros in 2012. Mortality from Comm/Mat/Per/Nutr was lowest in Nelson Mandela Bay and more than fivefold higher in Mangaung (5.63), more than threefold higher in Ekurhuleni (GP) (3.54), and more than double in Buffalo City (EC) (2.51), eThekwini (KZN) (2.37), Johannesburg (GP) (2.62) and Tshwane (GP) (2.51). Mortality due to HIV and TB, NCDs and injuries displayed the least variation, with Buffalo City (EC) and Mangaung (FS) having the highest mortality (more than double that of Nelson Mandela Bay (EC) for these cause groups.

Nbdcodename District

Death Year

2008 2012

0 1 2 3 4 5 6Comparative ASR

0 1 2 3 4 5 6Comparative ASR

Comm/Mat/Peri/Nutr N Mandela Bay: NMA

Johannesburg: JHB

Tshwane: TSH

Cape Town: CPT

eThekwini: ETH

Ekurhuleni: EKU

Buffalo City: BUF

Mangaung: MAN

HIV/AIDS and TB N Mandela Bay: NMA

Johannesburg: JHB

Tshwane: TSH

Cape Town: CPT

eThekwini: ETH

Ekurhuleni: EKU

Buffalo City: BUF

Mangaung: MAN

Non-communicable N Mandela Bay: NMA

Johannesburg: JHB

Tshwane: TSH

Cape Town: CPT

eThekwini: ETH

Ekurhuleni: EKU

Buffalo City: BUF

Mangaung: MAN

Injuries N Mandela Bay: NMA

Johannesburg: JHB

Tshwane: TSH

Cape Town: CPT

eThekwini: ETH

Ekurhuleni: EKU

Buffalo City: BUF

Mangaung: MAN

All causes N Mandela Bay: NMA

Johannesburg: JHB

Tshwane: TSH

Cape Town: CPT

eThekwini: ETH

Ekurhuleni: EKU

Buffalo City: BUF

Mangaung: MAN

2.75

0.48

1.17

1.37

1.36

1.00

1.22

1.82

2.37

5.63

1.00

3.54

2.62

2.51

1.41

2.51

0.95

1.49

0.53

0.47

1.00

0.70

0.44

1.24

0.95

2.28

2.53

1.46

1.00

1.00

0.90

1.51

0.75

1.48

0.67

0.86

1.00

0.80

1.20

0.84

1.27

1.57

2.37

2.23

1.00

1.40

1.40

1.32

0.85

1.18

0.77

1.00

0.84

1.04

1.52

0.81 1.45

1.28

2.29

2.33

1.00

1.80

1.80

1.32

0.75

0.95

0.68

1.47

1.00

0.84

1.42

0.81

1.55

2.58

1.33

2.36

1.00

1.44

1.54

1.41

Comparative age-standardised mortality ratios by metro

The Comparative ASR uses the lowest all cause age-standardised mortality rate for the latest year as the comparator (in this case N Mandela Bay in 2012).

251


Figure 14: Comparative age-standardised mortality ratios by metro, 2008 and 2012 (interpret with caution)

Figure 15 shows the cause of death profile in the metros based on the crudey and age-standardised mortality rates. Cape Town had the highest proportion of injury and NCD YLLs across all metros (Figure 13), yet the age-standardised mortality rate for injuries and NCDs was not the highest among the metros (Figure 14). In contrast, Mangaung (FS) had the highest proportions of YLLs due to Comm/Mat/Peri/Nutr and HIV and TB (Figure 13), and the highest age-standardised death rates for these cause groups (Figure 14). Comparative age-standardised mortality rates for all broad causes were highest in Mangaung (FS) and Buffalo City (EC) in 2008 and 2012. NCD age-standardised rates were more than double the age-standardised rates for Nelson Mandela Bay (EC) for males. Gender differentials were greatest for injury death rates, with male-to-female rate ratios ranging from 2.6 in Tshwane (GP) to 4.0 in Cape Town (WC) (Figure 16). However, in Mangaung (FS) and Buffalo City (EC), the gender differentials were huge across all causes. This suggests a possible problem, with population estimates for males being too low in those metros, and further investigation is needed. A dramatic decline in deaths recorded in Nelson Mandela Bay (EC) was also noted after 2010, which may reflect a data management or coding problem rather than a real reduction in mortality. The resulting age-standardised rates for Nelson Mandela Bay may therefore set an artificially low reference value for the comparative age-standardised rates by metro, complicating the interpretation of these results.

y Crude mortality represents the actual mortality burden experienced, while the age-standardised mortality rate is a weighted average of the age-specific mortality rates per 100 000 persons, where the weights are the proportions of persons in the corresponding age groups of the WHO standard population. YLLs represent the premature mortality (mortality occurring at younger ages which should be targeted for prevention).

District

Death Year

2012

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700Deaths per 100 000

N Mandela Bay:NMA

CMR

ASR

Johannesburg:JHB

CMR

ASR

Tshwane: TSH CMR

ASR

Cape Town:CPT

CMR

ASR

eThekwini: ETH CMR

ASR

Ekurhuleni: EKU CMR

ASR

Buffalo City:BUF

CMR

ASR

Mangaung: MAN CMR

ASR

317.5

129.8 420.3

127.0 50.5

53.5

50.7

55.2

123.8144.8

120.4 131.2 339.1

535.1

66.8

70.6

588.9

133.9

129.9

116.9 384.4

138.4

65.0

68.4

473.9

116.6

118.0

660.2

94.9

77.9

66.3

96.2

404.9

589.9

194.8

195.9

114.8

130.9

93.4

96.1

202.5

555.8189.8

170.4

195.4

377.1 77.6

77.6

138.8 122.3

585.3258.6

295.6

110.0

996.4

107.4

115.5615.5

937.5

298.5

328.5

258.3

124.6311.0

Crude mortality and age-standardised rates by metro, 2012

BroadcauseInjury

NCD

HIV and TB

Comm_mat_peri_nut

District Gender

Death Year

2012

0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400Deaths per 100 000

N Mandela Bay:NMA

Female

Male

Johannesburg:JHB

Female

Male

Tshwane: TSH Female

Male

Cape Town:CPT

Female

Male

eThekwini: ETH Female

Male

Ekurhuleni: EKU Female

Male

Buffalo City:BUF

Female

Male

Mangaung: MAN Female

Male

67.1

47.6

155.8

113.4

514.3

364.3

85.5

25.5

143.0

144.4

138.4

112.0

613.9

479.5

106.7

35.9

141.7

135.8

147.4

115.0

652.4

539.3

98.4

38.5

82.7

73.0

134.3

101.8

750.4

589.7

154.3

38.9

134.8

127.7

218.9

176.5

627.5

559.8

146.3

46.3

191.6

198.2

198.2

182.3

568.0

543.7

113.1

38.5

179.5

122.3

386.0

251.3

1,425.2

813.5

206.2

58.6

387.5

265.7

436.9

258.3

1,253.7

772.4

212.8

58.0

Age-standardised mortality rates by gender, by metro, 2012

BroadcauseInjury

NCD

HIV and TB

Comm_mat_peri_nut

252


Figure 15: Age-standardised and crude mortality rates by metro, 2012

Figure 16: Age-standardised mortality rates by gender, by metro, 2012

Figure 17 shows the 10 causes with the highest age-standardised mortality rates for each metro. High rates of mortality from TB, lower respiratory infection and HIV and AIDS featured in all metros, with extremely high TB mortality rates in Buffalo City (EC) (196.8 per 100 000) and Mangaung (FS) (184.7 per 100 000). Mangaung had the highest mortality rates for lower respiratory infection (167.1 per 100 000) and HIV and AIDS (143.8 per 100 000). Cardiovascular diseases and diabetes featured in all metros, with mortality rates for ischaemic heart disease higher than rates for cerebrovascular disease in Cape Town (WC), eThekwini (KZN), Johannesburg and Tshwane (GP), and cerebrovascular disease rates higher than rates for ischaemic heart disease in Buffalo City (EC), Ekurhuleni (GP), Mangaung (FS) and Nelson Mandela Bay (EC), suggesting that urban populations are at different stages in the health transition. COPD and oesophageal cancer mortality rates were very high in Buffalo City, and lung cancer mortality rates featured in Cape Town.

Compared with other metros, Buffalo City and Mangaung had very high mortality rates for most leading causes of death, particularly TB, cerebrovascular disease and HIV and AIDS. As suggested earlier, this may reflect population estimates that are too low for these two metros. Other possibilities could be that health services are suboptimal in these metros, or that these metros are heavily burdened as referral centres for severely ill patients from their surrounding areas.

253


Trends in age-standardised mortality rates by metro

All-cause age-standardised mortality rates declined in all metros between 2008 and 2013. In Buffalo City (EC), diarrhoeal disease dropped out of the top 10, lower respiratory tract infection declined, while HIV and AIDS, diabetes and hypertensive heart disease moved up in the ranking (Figure 18). In Cape Town (WC), HIV and AIDS, cerebrovascular disease and COPD, moved up in the ranking, while TB and lower respiratory disease moved down. In Ekurhuleni (GP), HIV and AIDS, cerebrovascular disease, ischaemic heart disease, hypertensive heart disease, diabetes mellitus and COPD moved up, while lower respiratory infections, TB and diarrhoea moved down. The ranking changed very slightly in eThekwini (KZN), with ischaemic heart disease moving above TB to the top. In Johannesburg (GP), HIV and AIDS, ischaemic heart disease, diabetes mellitus and COPD moved up in the ranking, while TB, lower respiratory infections and diarrhoea moved down. In Mangaung (FS), cerebrovascular disease and diabetes mellitus moved up in the ranking and HIV and AIDS and diarrhoea moved down. In Nelson Mandela Bay (EC), cerebrovascular disease moved above ischaemic heart disease. HIV and AIDS and diabetes moved up, while lower respiratory infection moved down. COPD moved up into the top 10. In Tshwane (GP), cerebrovascular disease, hypertensive heart disease, HIV and AIDS and diabetes mellitus moved up, while TB, diarrhoea and lower respiratory infections moved down.

District Rank.. Nbdcodename

0 20 40 60 80 100 120 140 160 180 200Deaths per 100 000

Buffalo City: BUF 1 Tuberculosis2 Cerebrovascular disease3 HIV/AIDS4 Ischaemic heart disease5 Hypertensive heart disease6 COPD7 Diabetes mellitus8 Lower respiratory infections9 Oesophagus10 Cardiomyopathy

Cape Town: CPT 1 Ischaemic heart disease2 Cerebrovascular disease3 HIV/AIDS4 Tuberculosis5 Trachea/bronchi/lung6 Diabetes mellitus7 COPD8 Lower respiratory infections9 Hypertensive heart disease10 Nephritis/nephrosis

Ekurhuleni: EKU 1 HIV/AIDS2 Lower respiratory infections3 Tuberculosis4 Cerebrovascular disease5 Ischaemic heart disease6 Hypertensive heart disease7 Diabetes mellitus8 COPD9 Diarrhoeal diseases10 Nephritis/nephrosis

eThekwini: ETH 1 Ischaemic heart disease2 Tuberculosis3 Cerebrovascular disease4 HIV/AIDS5 Lower respiratory infections6 Diabetes mellitus7 Nephritis/nephrosis8 Hypertensive heart disease9 Diarrhoeal diseases10 Mechanical forces

Johannesburg: JHB 1 HIV/AIDS2 Ischaemic heart disease3 Lower respiratory infections4 Cerebrovascular disease5 Tuberculosis6 Nephritis/nephrosis7 COPD8 Diabetes mellitus9 Hypertensive heart disease10 Septicaemia

Mangaung: MAN 1 Tuberculosis2 Lower respiratory infections3 Cerebrovascular disease4 HIV/AIDS5 Hypertensive heart disease6 Ischaemic heart disease7 Nephritis/nephrosis8 Diabetes mellitus9 Diarrhoeal diseases10 COPD

N Mandela Bay: NMA 1 Tuberculosis2 Cerebrovascular disease3 HIV/AIDS4 Ischaemic heart disease5 Hypertensive heart disease6 Diabetes mellitus7 Nephritis/nephrosis8 Trachea/bronchi/lung9 COPD10 Lower respiratory infections

Tshwane: TSH 1 Ischaemic heart disease2 Cerebrovascular disease3 Hypertensive heart disease4 HIV/AIDS5 Lower respiratory infections6 Tuberculosis7 Diabetes mellitus8 Nephritis/nephrosis9 Diarrhoeal diseases10 COPD

196.8154.3

98.888.8

71.765.965.6

60.157.6

40.7108.3

79.764.8

51.847.2

39.539.3

31.931.931.5

102.591.6

87.377.777.5

52.839.6

36.528.9

26.9110.7

109.894.8

86.159.7

48.142.6

37.828.2

26.169.4

64.362.9

59.054.4

36.332.0

28.327.6

23.6184.7

167.1160.1

143.895.3

90.052.7

50.545.2

42.772.8

64.157.0

45.837.2

30.922.4

21.120.5

18.276.9

75.869.969.0

67.660.9

45.333.3

27.225.6

Leading age-standardised mortality rates by metro, 2012

254


Figure 17: Leading age-standardised mortality rates by metro, 2012

District

2007 2008 2009 2010 2011 2012 2013 2014Year of death

Buffalo City:BUF

Cape Town:CPT

Ekurhuleni:EKU

eThekwini:ETH

Johannesburg:JHB

Mangaung:MAN

N MandelaBay: NMA

Tshwane: TSH

Cardiomyopathy (10)Nephritis/nephrosis (10)

Diarrhoeal diseases (10)Oesophagus (10)

Oesophagus (9)Cardiomyopathy (9)

COPD (5)

COPD (6)

Hypertensive heart disease (7)


Diarrhoeal diseases (8)

Diabetes mellitus (9)


HIV/AIDS (6)

HIV/AIDS (3)

Lower respiratory infections (4)


Ischaemic heart disease (3)


Cerebrovascular disease (2) Cerebrovascular disease (2)

Tuberculosis (1) Tuberculosis (1)

Trachea/bronchi/lung (5) Trachea/bronchi/lung (5)

Nephritis/nephrosis (10) Nephritis/nephrosis (10)

COPD (8)

COPD (7)

Hypertensive heart disease (9) Hypertensive heart disease (9)

Diabetes mellitus (6) Diabetes mellitus (6)

HIV/AIDS (4)

HIV/AIDS (3)



Ischaemic heart disease (1) Ischaemic heart disease (1)

Cerebrovascular disease (3)

Cerebrovascular disease (2)Tuberculosis (2)

Tuberculosis (4)

Mechanical forces (9)

Nephritis/nephrosis (10)Nephritis/nephrosis (10)COPD (10)

COPD (8)







HIV/AIDS (3)

HIV/AIDS (1)Lower respiratory infections (1)



Ischaemic heart disease (5)Cerebrovascular disease (5)


Tuberculosis (2)

Tuberculosis (3)



Nephritis/nephrosis (7) Nephritis/nephrosis (7)







HIV/AIDS (4) HIV/AIDS (4)

Lower respiratory infections (5) Lower respiratory infections (5)



Cerebrovascular disease (3) Cerebrovascular disease (3)

Tuberculosis (1)

Tuberculosis (2)

Mechanical forces (10) Septicaemia (10) Septicaemia (10)

Nephritis/nephrosis (8)


COPD (10)

COPD (7)Hypertensive heart disease (7)






HIV/AIDS (3)

HIV/AIDS (1)Lower respiratory infections (1)






Tuberculosis (2)

Tuberculosis (5)



COPD (10) COPD (10)





Diabetes mellitus (8) Diabetes mellitus (8)

HIV/AIDS (3)

HIV/AIDS (4)

Lower respiratory infections (2) Lower respiratory infections (2)






COPD (9)


Asthma (8)

Asthma (9)Trachea/bronchi/lung (9)

Trachea/bronchi/lung (8)Nephritis/nephrosis (8)


COPD (9)





HIV/AIDS (5)

HIV/AIDS (3)








Road injuries (10)Nephritis/nephrosis (9)


COPD (10) COPD (10)







HIV/AIDS (6)

HIV/AIDS (4)



Ischaemic heart disease (1) Ischaemic heart disease (1)



Tuberculosis (3)

Tuberculosis (6)

Trends in leading age-standardised mortality rates (rank) by metro

Max. BroadcauseComm_mat_peri_nutHIV and TBInjuryNCD

ASR18.2

100.0200.0300.0355.2

255


Figure 18: Trends in leading age-standardised mortality rates by metro, 2008-2012

256


Conclusion

Mortality rates in South Africa declined between 2008 and 2013, mainly due to a decline in HIV-related mortality. Despite this, HIV and AIDS and associated conditions still stand out as being a leading cause of YLLs, together with cerebrovascular diseases, ischaemic heart disease, diabetes mellitus, road injuries, interpersonal violence and hypertensive heart disease.

A reduction in the percentage of deaths coded to ill-defined causes or garbage codes was noted in the Western Cape. This suggests that the Western Cape local mortality surveillance system, which included a provincial training initiative in medical certification of cause of death as well as increased utilisation of mortality information for health policy making, may have had a positive impact on the quality of medical certification. However, until the completeness of death registration is consistently high across all districts and the quality of medical certification has improved, the district-level mortality profiles need to be interpreted cautiously. In particular, lack of reliability of the injury profile and the misclassification of HIV and AIDS need to be taken into consideration. Efforts to utilise mortality profile information at district level need to be accompanied by initiatives to improve medical certification of the cause of death as well as the geographical coding of place of residence and place of death. Urgent initiatives are required to improve the quality of injury mortality information in the national statistics as these are currently misleading.

13 Burden of disease Pam Groenewald, Debbie Bradshaw ... Health Barometers/13... · n Dorrington R, Bradshaw D. Maternal mortality in South Africa: lessons from a case study in the

Documents