Process Indicators

��

Nurse clinical workload, 2005/06

Legend

NURSE_05no data

less than 20

20 - 29

30 - 39

40 - 49

over 50


Legend

NURSE_05no data

less than 20

20 - 29

30 - 39

40 - 49

over 50


Legend

NURSE_05no data

less than 20

20 - 29

30 - 39

40 - 49

over 50

SectionA:IndicatorcomparisonsbyDistrict

�. ProcessIndicators

�.� NurseclinicalWorkload

The nurse clinical workload measures the average daily number of patients attended to by a professional nurse in PHC facilities. This is a measure of efficiency and very low values indicate that scarce skills (i.e. professional nurse time) are not being optimally utilised. Very high values indicate that either the data is incorrect or that nurses are seeing too many patients per day with resultant compromise in quality or burn-out, or both.

District View In South Africa in 2005 the nurse clinical workload averages 3�.6 patients a day, almost exactly the same as in 2004. However there are a number of districts with incorrect or missing data which affect this average. In particular none of the 6 districts in the Western Cape was able to supply this indicator.

Figure �6 shows the range from �2.9 in Waterberg (LP) to an improbable 374.2 in iLembe (KZN). Clearly the data in iLembe, as well as that in Pixley ka Seme (NC) of �40.7 and Namakwa (NC) of �33.4 are incorrect.2

Four of the five highest workloads were from districts in the Northern Cape. It is possible that the way in which the data is being collected or the definitions used are contributing to these excessively high rates. Further investigation into these data is required.

Limpopo was responsible for four of the five lowest workload indicators with values of less than �5 patients a day in these districts. This (being an average result for the district) also requires investigation as it implies that many professional nurses are seeing less than �0 patients a day.

Data has not been available for any of the districts in the WC since 2003 and it is important for nationally accepted indicators to be supplied by all districts.

Generally the districts from particular provinces were more closely clustered than in 2004. This implies either improved consistency in the data, or a greater move to equity with staff increases in busier facilities, or both.

Map 3: Nurse clinical workload in South Africa, 2005/06

� ThehighvaluesiniLembe,Namakwa,KgalagadiandPixleykaSemearemissingdenominatorsforalmostalldata.Insomecasestherearealsomonthsmissingdataaltogether.Thismeansthereisheadcountdata,butdataonclinicalworkdaysinmissing.


0 10 20 30 40 50 60 70 80 90 100

Cape WinelandsCentral Karoo

City of Cape TownEden

OverbergWest CoastWaterbergCapricorn

Greater SekhukhuneAmajuba

BohlabelaSisonke

City of JohannesburgAlfred Nzo

UmzinyathiUgu

UmkhanyakudeChris Hani

MopaniO.R. Tambo

ZululandUthukela

AmatholeUkhahlamba

VhembeXhariepMotheo

UthunguluMetsweding

UMgungundlovuEkurhulenieThekwini

LejweleputswaThabo Mofutsanyane

West RandSedibeng

Fezile DabiCentral

City of TshwaneNkangala

CacaduBojanala

Nelson Mandela Bay MetroEhlanzeniKgalagadi

Gert SibandeBophirimaSouthern

Frances BaardSiyanda

NamakwaPixley ka Seme

iLembeSouth Africa

Patients per nurse clinical work day

ECFSGPKZNLPMPNCNWWCSA

374141133

no data

��


Figure 16: Nurse clinical workload by district, 2005/06


0 10 20 30 40 50 60 70 80 90 100

Central KarooGreater Sekhukhune

BohlabelaAlfred Nzo

UmzinyathiUgu


O.R. TamboZululand

UkhahlambaThabo Mofutsanyane

KgalagadiISRDP average

South Africa



no data


0 10 20 30 40 50 60 70 80 90 100

City of Cape Town

City of Johannesburg

Ekurhuleni

eThekwini

City of Tshwane

Nelson Mandela Bay Metro

Metro average

South Africa



no data


0 10 20 30 40 50 60 70 80 90 100

Central KarooGreater Sekhukhune

BohlabelaAlfred Nzo

UmzinyathiUgu


O.R. TamboZululand

UkhahlambaThabo Mofutsanyane

KgalagadiISRDP average

South Africa



no data


0 10 20 30 40 50 60 70 80 90 100

City of Cape Town


Ekurhuleni

eThekwini

City of Tshwane


Metro average

South Africa



no data

��


Metro View The metropoles have an average clinical nurse workload of 39.3, higher than that of the national average. Workloads vary from �7.7 in Johannesburg to 48.4 in Nelson Mandela metro.

Figure 17: Nurse clinical workload by metro district, 2005/06

Rural Nodes Only two rural node districts reported a nurse clinical workload above the South African average of 32 patients per professional nurse per day and only one district reported a workload greater than the metropolitan district average of 39 patients per day. The workload was the highest in Kgalagadi with a workload of 50 patients per day and the lowest in Greater Sekhukhune with an average �3 patients per day.

The relatively low nurse clinical workload in rural node districts may be due to relative overstaffing, or it may be due to the low population density with longer times taken to do outreach activities and difficulties in travelling and transport in rural areas.

Figure 18: Nurse clinical workload by rural node, 2005/06

��

Average length of stay, 2005/6

Legend

ALOS_05no data

< 3 days

3 - 3.9

4 - 4.9

5 - 5.9

6 days and over


�.� AverageLengthofStay

The average length of stay (ALOS) measures how long patients spend in district hospitals. It is calculated by dividing the number of patient days by the number of separations, which include transfers, discharges and deaths. District hospitals generally admit acute, relatively uncomplicated patients and the idea is to treat them and discharge them as soon as is possible.

The ALOS is a proxy measure for the quality of care received as well as of the efficiency of the hospital. The guideline figure from the National DOH is and ALOS of 3.4 days for district hospitals.

District View As can be seen in Map 4 and Figure �9, there was a wide variation in the ALOS in district hospitals throughout South Africa in 2005. The average ALOS for the country was 4.3 days. The figures range from 2.2 days in Pixley ka Seme (NC) to 9.� in Chris Hani (EC). Because many of the districts contain more than one district hospital, the district averages seen in this graph conceal the greater individual hospital variations. There was no data from Metsweding (GP), as this district does not have district hospitals. There was also no data from Bohlabela, as the data from this district has been incorporated into neighbouring districts.3 The �0 districts with the highest ALOS were districts in the Eastern Cape and KwaZulu-Natal provinces.

There were distinct differences across the provinces. All the districts in the Northern Cape, Western Cape and Mpumalanga provinces had an ALOS of less than four days while six of the seven Eastern Cape districts and nine of the eleven KwaZulu-Natal districts had an ALOS of greater than 4 days.

There were generally large intra-provincial variations with a wide range between the district with the lowest and the district with the highest ALOS. Exceptions to this were the Western Cape where the districts are fairly tightly clustered around an ALOS of three days and Limpopo province where the clustering is around four days ALOS.

The reasons for both the intra and inter-provincial differences should be investigated by managers at the provincial and national Departments of Health.

Map 4: Average length of stay in district hospitals in South Africa 2005/06

� AsdetailedinthegovernmentgazetteNr�8�6�of��December�005,thecross-boundarymunicipalityofBohlabela,hasbeendividedbetweenLimpopoandMpumalangaprovincesandthusno longerexists.TheDecember�006DHISdatasethas incorporated thedata forBohlabela intoMpumalangaandLimpopodistrictsandthereforeanydistricthospitalsthatwerepreviouslyclassifiedunderBohlabelo, fall intoeitheroneofthoseprovinces.


0 2 4 6 8 10

South AfricaBohlabela

MetswedingPixley ka Seme

XhariepFezile Dabi

NkangalaOverberg

West CoastFrances Baard

City of TshwaneLejweleputswa

SouthernCape Winelands

City of Cape TownEden

SedibengThabo Mofutsanyane

EkurhuleniAmajuba

KgalagadiSiyanda

West RandNelson Mandela Bay

Central KarooCentral

EhlanzeniUMgungundlovu

MopaniGreater Sekhukhune

NamakwaBojanala

Gert SibandeCacadu

WaterbergCapricornBophirima

UthukelaCity of Johannesburg

SisonkeMotheo

eThekwiniVhembe

UkhahlambaZululand

UguChris HaniUmzinyathi

AmatholeUthungulu

UmkhanyakudeiLembe

Alfred NzoO.R. Tambo

Days


No district hospitals

�0


Figure 19: Average length of stay by district 2005/06

Metro View As can be seen in Figure 20, the average for the metros of 3.6 days was less than the average for all district hospitals in South Africa. The figures range from 2.5 days in Tshwane to 4.9 days in Nelson Mandela. Two of the six metros (Johannesburg and eThekwini) had an ALOS greater than the South African average. The reasons for such long average stays in these two urban districts needs further investigation at individual hospital level.


0 2 4 6 8 10

South AfricaISRDP average

BohlabelaThabo Mofutsanyane

KgalagadiCentral Karoo

Greater SekhukhuneUkhahlamba

ZululandUgu

Chris HaniUmzinyathi

UmkhanyakudeAlfred Nzo

O.R. Tambo

Days



0 2 4 6 8 10

South Africa

Metro average

City of Tshwane

City of Cape Town

Ekurhuleni



eThekwini

Days



0 2 4 6 8 10

South AfricaISRDP average

BohlabelaThabo Mofutsanyane

KgalagadiCentral Karoo

Greater SekhukhuneUkhahlamba

ZululandUgu

Chris HaniUmzinyathi

UmkhanyakudeAlfred Nzo

O.R. Tambo

Days



0 2 4 6 8 10

South Africa

Metro average

City of Tshwane

City of Cape Town

Ekurhuleni



eThekwini

Days


��


Figure 20: Average length of stay by metro area 2005/06

Rural View The ALOS in the rural nodes of 5.5 days was greater than the SA average. There were two distinct patterns within the rural districts. All of the eight rural districts in the Eastern Cape and KwaZulu-Natal provinces had an ALOS of more than 5 days while the rural districts in the other provinces had an ALOS of less than 4 days. The KwaZulu-Natal districts were clustered in a tight range between 6.� and 7.2 days.

The situation in O.R. Tambo (EC), with an ALOS of 9.� days and in Alfred Nzo (EC) with an ALOS of 8.7 days, both of which are far higher than the next highest district (Umkhanyakude in KwaZulu-Natal with an ALOS of 7.2 days), needs urgent managerial investigation as to why patients are spending such long periods in the individual hospitals in these districts. One possible explanation for the long ALOS in the rural hospitals generally may be due to the shortage of doctors who can discharge or transfer patients timeously.

Figure 21: Average length of stay in the rural nodes

Change in average length of stay, 2003/04 - 2005/06

-3 -2 -1 1 2

South AfricaUmkhanyakude

CapricornNelson Mandela Bay

iLembeVhembe

eThekwiniAmajubaUthukela

MopaniO.R. Tambo

UthunguluUMgungundlovu

SiyandaGreater Sekhukhune

SisonkeFrances Baard

ZululandUgu

Alfred NzoEkurhuleniNkangala

CentralPixley ka Seme

AmatholeWest RandWest Coast

KgalagadiChris HaniWaterberg

XhariepEden

Fezile DabiUmzinyathi

Gert SibandeLejweleputswa

OverbergBohlabela

MetswedingCity of Tshwane

CacaduCape Winelands

SouthernMotheo

UkhahlambaBojanala

City of JohannesburgBophirima

Thabo MofutsanyaneCentral Karoo

SedibengCity of Cape Town

EhlanzeniNamakwa

Days



��


Change in average length of stay

The graph comparing the change in ALOS between 2003 and 2005, shows that there was an improvement of 0.3 days across the country (see Figure 22). Twelve of the districts showed an increase in the ALOS, whilst the remainder either stayed the same (3 districts) or had a decrease in the ALOS (36 districts).

All eleven KwaZulu-Natal districts improved their efficiency and decreased the ALOS, as did the districts in Limpopo.

Figure 22: Change in average length of stay by district 2003/04 – 2005/06

��

Usable bed utilisation rate, 2005/6

Legend

BUR_05no data

< 50%

50 - 59%

60 - 69%

70 - 79%

80% and over


�.� BedUtilisationRate

Bed utilisation (occupancy) rate is a measure of the occupancy of the beds available for use. It is calculated by dividing the number of patient days by the bed days available, over a specific time period (usually a year), and expressing this as a percentage. It is generally a measure of efficiency and expresses how well the hospital is using its available capacity. The indicative value set by the national DOH is 72%.

The bed days available in the numerator of the calculation need to be calculated correctly in order to accurately reflect the number of beds available. South African hospitals classify beds as “authorised”, (this number reflecting the number of beds that “should” be in use), and “actual”, those that are really being used. When the actual numbers of beds in use changes, this needs to be reflected in the number of bed days calculation.

District View In Figure 23 and Map 5, the variation in the bed utilisation rate (BUR) across the districts can clearly be seen. The average BUR for South Africa in 2005 was 63.9%. There is a wide range in the BURs from a low of 36.7% in Amajuba (KZN) to a high of 83.�% in Central Karoo (WC). Because many of the districts contain more than one district hospital, the district averages seen in this graph conceal greater individual hospital variations. There was no data from Metsweding (GP) as this district does not have district hospitals. There was also no data from Bohlabela as the data from this district has been incorporated into neighbouring districts.4

Three districts had BURs of less than 50% and a further �0 districts had BURs below 60%. This points to large scale inefficiencies in the district hospital system and suggests that provinces should ensure that their strategic transformation plans include changing under-utilised hospitals into community health centres.

There were wide variations among districts within provinces. As with the ALOS indicator, many districts have more than one district hospital and the individual hospital differences have been decreased by using district averages.

Map 5: Usable bed utilisation rate in South Africa 2005/06

� AsdetailedinthegovernmentgazetteNr�8�6�of��December�005,thecross-boundarymunicipalityofBohlabela,hasbeendividedbetweenLimpopoandMpumalangaprovincesandthusno longerexists.TheDecember�006DHISdatasethas incorporated thedata forBohlabela intoMpumalangaandLimpopodistrictsandthereforeanydistricthospitalsthatwerepreviouslyclassifiedunderBohlabelo, fall intoeitheroneofthoseprovinces.

Bed utilisation rate, 2005/06

0 10 20 30 40 50 60 70 80 90 100

BohlabelaMetsweding

AmajubaO.R. Tambo

CentralUthungulu

ZululandBojanala

Pixley ka SemeFrances Baard

UthukelaEhlanzeniOverberg

WaterbergNkangala

SisonkeiLembe

AmatholeThabo Mofutsanyane

UguCapricornSouthern

MopaniGreater Sekhukhune

SedibengChris Hani

City of TshwaneLejweleputswa

UmkhanyakudeCity of

Alfred NzoWest CoastWest Rand

UMgungundlovuUmzinyathiBophirima

CacaduVhembe

eThekwiniEkurhuleni

City of Cape TownUkhahlamba

Fezile DabiGert Sibande

XhariepCape Winelands

KgalagadiNelson Mandela Bay

EdenSiyandaMotheo

NamakwaCentral Karoo

South Africa

Percentage



��


Figure 23: Usable bed utilisation rate by district 2005/06

Metro View As one would expect, with less district hospital beds available as compared to rural areas, the metro BUR of 7�.6% is greater than the SA average and is in line with the indicative value of 72% set by the national Department of Health. There is a tight range from 66.7% in Tshwane to 76.8% in Nelson Mandela.


0 10 20 30 40 50 60 70 80 90 100

BohlabelaO.R. Tambo

ZululandThabo

UguGreater

Chris HaniUmkhanyaku

Alfred NzoUmzinyathi

UkhahlambaKgalagadi

Central KarooISRDP

South Africa

Percentage



0 10 20 30 40 50 60 70 80 90 100

City of Tshwane


eThekwini

Ekurhuleni

City of Cape Town


Metro average

South Africa

Percentage



0 10 20 30 40 50 60 70 80 90 100

BohlabelaO.R. Tambo

ZululandThabo

UguGreater

Chris HaniUmkhanyaku

Alfred NzoUmzinyathi

UkhahlambaKgalagadi

Central KarooISRDP

South Africa

Percentage



0 10 20 30 40 50 60 70 80 90 100

City of Tshwane


eThekwini

Ekurhuleni

City of Cape Town


Metro average

South Africa

Percentage


��


Figure 24: Usable bed utilisation rate 2005/06 in the metro districts

Rural Nodes The BUR for the rural districts was 62.5%, which is very close to the SA average of 63.9%. There was a very wide range from a high BUR in Central Karoo of 83.�% to a low of 48% in O.R. Tambo. Three districts had a BUR of greater than 70% while two had a BUR of less than 60%. It was concerning that O.R. Tambo, which had the highest ALOS also had the second lowest BUR in South Africa. These indicators reflect a dysfunctional hospital system in this district.

Figure 25: Usable bed utilisation rate in the rural nodes 2005/06

Change in bed utilisation rate, 2003/04 - 2005/06

-30 -20 -10 0 10 20 30 40

Pixley ka SemeNelson Mandela Bay

SiyandaAlfred Nzo

West RandSouthern

EkurhuleniUkhahlamba

EhlanzeniUthungulu

Fezile DabiFrances BaardLejweleputswa

BojanalaCapricorn

MetswedingBohlabelaUthukela

City of TshwaneCentral

AmajubaMopani

OverbergSisonke

O.R. TamboZululand

AmatholeVhembe

Thabo MofutsanyaneSouth Africa

NkangalaWaterberg

iLembeWest Coast

CacaduGert Sibande

MotheoUMgungundlovuCape Winelands

BophirimaSedibeng

Chris HaniUgu

eThekwiniUmkhanyakude

Greater SekhukhuneEden

XhariepUmzinyathiKgalagadi

City of JohannesburgCity of Cape Town

Central KarooNamakwa

South Africa

Percentage



��


Change in usable bed utilisation rate

Figure 26 shows the change in the BUR between 2003 and 2005. During this period the average BUR for South Africa increased by 3.3. In total, fifteen districts had a decreased BUR and of these, four decreased by more than �0%. In 36 districts there was an increased BUR, ten of which increased their BURs by �0% or more.

There were wide swings in the Northern Cape districts with Namakwa having the largest increase in South Africa of 38.8% and Pixley ka Seme having the largest decrease in South Africa of 24.2%. These large changes suggest some degree of data unreliability in the Northern Cape.

Figure 26: Change in bed utilisation rate by district, 2003/04 - 2005/06

Documents

Process Indicators