Russell et al.

Human Resources for Health (2017) 15:52

DOI 10.1186/s12960-017-0229-9

RESEARCH Open Access

Patterns of resident health workforce

turnover and retention in remote
communities of the Northern Territory
of Australia, 2013–2015
Deborah J Russell1*, Yuejen Zhao2, Steven Guthridge2, Mark Ramjan2, Michael P Jones3, John S Humphreys4
and John Wakerman5

Abstract
Background: The geographical maldistribution of the health workforce is a persisting global issue linked to
inequitable access to health services and poorer health outcomes for rural and remote populations. In the Northern
Territory (NT), anecdotal reports suggest that the primary care workforce in remote Aboriginal communities is
characterised by high turnover, low stability and high use of temporary staffing; however, there is a lack of reliable
information to guide workforce policy improvements. This study quantifies current turnover and retention in remote
NT communities and investigates correlations between turnover and retention metrics and health service/community
characteristics.
Methods: This study used the NT Department of Health 2013–2015 payroll and financial datasets for resident health
workforce in 53 remote primary care clinics. Main outcome measures include annual turnover rates, annual stability
rates, 12-month survival probabilities and median survival.
Results: At any time point, the clinics had a median of 2.0 nurses, 0.6 Aboriginal health practitioners (AHPs), 2.2 other
employees and 0.4 additional agency-employed nurses.
Mean annual turnover rates for nurses and AHPs combined were extremely high, irrespective of whether turnover
was defined as no longer working in any remote clinic (66%) or no longer working at a specific remote clinic (128%).
Stability rates were low, and only 20% of nurses and AHPs remain working at a specific remote clinic 12 months after
commencing. Half left within 4 months.
Nurse and AHP turnover correlated with other workforce measures. However, there was little correlation between most
workforce metrics and health service characteristics.
Conclusions: NT Government-funded remote clinics are small, experience very high staff turnover and make
considerable use of agency nurses. These staffing patterns, also found in remote settings elsewhere in Australia and
globally, not only incur higher direct costs for service provision—and therefore may compromise long-term
sustainability—but also are almost certainly contributing to sub-optimal continuity of care, compromised health
outcomes and poorer levels of staff safety. To address these deficiencies, it is imperative that investments in
implementing, adequately resourcing and evaluating staffing models which stabilise the remote primary care
workforce occur as a matter of priority.
Keywords: Remote health, Rural workforce, Health workforce, Fly-in/fly-out, Rural health services, Aboriginal, Aboriginal
health practitioner, Remote area nurse, Turnover, Retention

* Correspondence: [Link]@[Link]
1
Monash Rural Health, Monash University, PO Box 666, Bendigo, Victoria
3552, Australia
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License ([Link] which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
([Link] applies to the data made available in this article, unless otherwise stated.
Russell et al. Human Resources for Health (2017) 15:52 Page 2 of 12

Background (NTER) which was initiated in 2007 [19]. While many

Geographical maldistribution of the health workforce is changes occurred in the targeted NT remote Aboriginal
a persisting global issue that has been linked with the in- communities as a result of NTER, one key feature
ability of rural and remote populations to gain equitable impacting on the health workforce in remote commu-
access to health services and consequent poorer health nities was the extra financial resources available, lead-
outcomes [1]. Maldistribution results in absolute and ing to increased use of short-term visiting health
relative shortages of rural primary care workers in many workers. The Remote Area Health Corps (RAHC), for
countries, irrespective of the country’s wealth [2–7]. In example, was established in 2008 and received Australian
Australia, the 2013 Review of Australian Government Government funding to recruit metropolitan-based health
Health Workforce Programs found that the most signifi- professionals for periods of service of 3 to 12 weeks [20].
cant primary care workforce issue was the distribution It has been suggested, however, that high levels of short-
of health workers [8]. Addressing inequities relating to term visiting health workers in NT Aboriginal communi-
the maldistribution of health workers requires workforce ties causes frustration for experienced permanent staff
planners to have a good understanding of health worker and may impact negatively on long-term service sustain-
transitions into and out of rural and remote areas. This, ability, quality of care and, ultimately, health outcomes.
in turn, requires measurement of these transitions and However, the extent to which quantitative evidence sup-
monitoring changes over time, so that the effectiveness ports such assertions has not yet been rigorously exam-
of any strategies devised to optimise health worker re- ined [21, 22].
cruitment, turnover and retention can be developed and Additionally, in 2002, the Aboriginal and Torres Strait
assessed. Islander Workforce National Strategic Framework pro-
Recent research suggests that no single workforce posed changes affecting the registration requirements,
metric assessing turnover or retention is sufficient, as professional recognition and role of Aboriginal health
each is likely to have considerable limitations if used in workers (AHWs). In the same year, completion of Cer-
isolation [9]. Instead, a suite of measures provides more tificate 4 competency-based training became a require-
comprehensive information about both the health ment for newly registered AHWs in NT and for
workers who are leaving rural and remote communities existing AHWs with Certificate 3 level qualifications
and, perhaps more importantly, those who are staying who wished to progress to a higher career level. Num-
[10, 11]. While the health workforce literature identifies bers of AHWs registered in the NT fluctuated between
many different indicators of turnover and of retention approximately 250 and 340 between 2004 and 2010, a
[12], key metrics relevant to the Australian rural and decline from registrations in excess of 400 in the late
remote health workforce context include annual turn- twentieth century [23]. In July 2012, national registra-
over rates, stability rates, survival probabilities and me- tion for Aboriginal and Torres Strait Islander health
dian survival [9]. A second component of workforce practitioners (hereafter referred to, using the NT term,
mobility is the level at which turnover or retention out- as Aboriginal health practitioners, or AHPs) was intro-
comes are determined and reported. This level has been duced, replacing the previous classification of AHWs.
variously defined, for example, retention in a practice, In 2012, 2013 and 2014, there were 219, 216 and 201
retention in a community, and retention in a larger registered AHPs in the NT respectively, representing a
health organisation [13]. Data from the United States of further decline in AHP numbers [24]. How these recent
America indicate annual health service turnover rates reductions in numbers of AHPs registered in NT have
for hospital-based nurses of up to 20% [14, 15]. Other manifested in remote NT Aboriginal communities, in
comparative research, also from the USA, suggests that terms of supply, turnover and stability of AHPs
annual hospital staff turnover rates of 4–12% are low, employed in a permanent capacity by NTG DoH is not
12–21% medium and 22–44% high [16]. Australian re- known.
search reports average annual organisational turnover Other gaps in knowledge include current turnover
of nurses from the Northern Territory Government rates, stability rates and survival probabilities for nurses
(NTG) Department of Health (DoH) of 35% and organisa- and for AHPs; correlation between key workforce met-
tional turnover of permanent Queensland Health nurses rics in these remote contexts; and whether there are
of 20%, with much higher turnover rates and lower stabil- differences in turnover rates, stability rates and health
ity rates experienced by nurses working in smaller more worker survival patterns amongst the different remote
remote health services [17, 18]. communities and the extent to which differences are
Published Northern Territory (NT) reports are now related to observable community characteristics such as
10 or more years old, and considerable changes have geographical remoteness, community population size
occurred in the ensuing period, particularly with regard and health service size. Finally, we do not know to what
to the Australian Government’s NT Emergency Response extent use of casual and agency nurse staffing in remote
Russell et al. Human Resources for Health (2017) 15:52 Page 3 of 12

communities is associated with turnover and retention remote clinics. Some nurses recruited through a nurse
of permanent employees and hence with workforce sus- employment agency are employed on temporary or cas-
tainability in remote communities. These are important ual NTG DoH contracts and are captured by the PIPS
evidence gaps that limit the ability of workforce plan- dataset. Other agency nurses are paid directly by an
ners and policymakers to understand and address sub- agency and are therefore not included in PIPS data.
optimal health service performance related to high These agency-employed nurses, however, are identifi-
turnover and poor retention of health workers in re- able through labour hire costs in NTG Government
mote communities. Accounting System (GAS). GAS expenditures on
The aims of this paper, therefore, are threefold. agency nurse labour hire costs were used to derive the
Firstly, to describe current (2013–2015) health work- aggregated full-time equivalent (FTE) agency-employed
force turnover and retention patterns for nurses and nurses working in remote health services using the
AHPs living and providing clinical services in govern- standard NTG DoH formula of agency labour expenses
ment health services in remote NT communities. divided by twice the departmental annual average nurse
Secondly, to investigate how key workforce metrics personnel cost [26].
correlate with each other. Thirdly, to investigate how
key workforce metrics correlate with health service
characteristics, including community remoteness and Analysis
isolation measures, health service population catch- In this study, an exit was primarily defined at a health
ment, and health service size. service level (primary turnover profile), that is, when an
employee left one of the 53 specific health clinics for a
Methods period of more than 12 weeks.
Study setting A secondary definition of an exit, at a remote health
This study is set in the remote NT of Australia, covering level (secondary turnover profile), is also reported, when
an area of approximately 1.3 million km2. The NT is an employee ceased working in remote health, that is, no
sparsely settled. Of the population of approximately longer worked in any of the 53 remote health clinics. In
244,000 people, 27% identify as Aboriginal. Most (80%) this instance, remote inter-clinic moves were ignored.
NT Aboriginal people live in remote towns and communi- Measures of health worker supply were number of
ties [25]. During the study period, the NTG DoH provided unique employees, average annual headcounts, average
health care services in 53 remote NT communities, while FTE, and agency-employed nurse FTE ratios, defined as
non-government organisations (Aboriginal Controlled follows:
Community Health Organisations) provided health care
services in a further 28 remote communities. Primary care 1. Total number of unique employees (sum of
delivery in remote NT communities is generally provided individuals employed in each year)
by resident Remote Area Nurses and Midwives (hereafter 2. Average annual headcounts (average number
referred to as nurses) and AHPs with professional sup- of individuals employed in each pay period in
port provided by telehealth and scheduled intermittent each year)
visits from medical and allied health practitioners. Two 3. Average FTE (average FTE employed in each pay
health services in very small remote communities did period in each year)
not have nurses, and 11 health services did not have 4. Agency-employed nurse FTE ratio
AHPs; however, there were no other systematic differ-
ences in health service characteristics. Primary care in
remote communities was supported by secondary care average agency employed nurse FTE
¼
available in five public NT hospitals. The focus of this sum of nurse FTE on organisational chart
study was on the resident clinical staff (nurses and
AHPs) in remote NTG-operated health services. The key turnover and retention metrics were averaged
over 3 years and based on headcounts (except where
specified as FTE):
Data
Two separate data sources were used. The first was the 1. The turnover measure was Annual turnover rates (%)
Personnel Information and Payroll Systems (PIPS) data
from NTG DoH (3 January 2013 to 30 December total number of exits in 12 month period
¼  100
2015). This provides comprehensive, individual-level, average number of employees in 12 month period
de-identified information on all nurses and AHPs paid
directly through the NTG DoH payroll in any of the 53 The retention measures were
Russell et al. Human Resources for Health (2017) 15:52 Page 4 of 12

2. Annual stability rates (%)

Number of employees at start of year who remain employed 12 months later

¼  100
Number of employees at start of year

3. Experienced nurses or AHPs

Number of pay periods per calendar year with at least one nurse or AHP
¼
who has been at that health service for 2 or more years continuously

4. Survival probability after 12 months (nurses and AHPs)

Number remaining employed beyond 12 months after commencing at a remote health service
¼
Number at risk of exiting a remote health service

5. Median survival time in years (nurses and AHPs)

Time from commencing at a remote health service at which the probability of remaining
¼
employed at the health service equals the probability of having exited which is equal to 0:5

Summary statistics for each of the key metrics were to explore associations between key workforce metrics
analysed by community population size and whether and also between key workforce metrics and health
the communities were predominantly Aboriginal or service characteristics.
not. Aboriginal community population size was dis- Ethics approval was received from the Human Research
tributed into four categories (<200, 200–349, 350–799, Ethics Committee of the NTG DoH and Menzies School
≥800). Since there were only seven predominantly of Health Research (2015-2363).
non-Aboriginal communities, it was not appropriate
to stratify non-Aboriginal communities by population
size. Community population size estimates were based Results
on 30 June 2012 Australian Bureau of Statistics Esti- In total, the number of unique nurses, AHPs and other
mated Resident Populations. 2011 Australian Bureau staff members on the payroll for the 53 clinics in the 3-
of Statistics Census data for Indigenous Locations year study period were 470, 93 and 583 respectively. The
were used to determine whether a community was average annual headcounts for nurses, AHPs and other
predominantly (>50%) Aboriginal or not. 2012 ABS employees were 272, 67 and 314 respectively. The total
Estimated Resident Population data were very strongly FTE for the 53 clinics was 68.7 and 23.8 for nurses and
(r = 0.97) correlated with 2014 health service records AHPs respectively. In any pay period, the median number
of catchment populations which clinic staff maintain of nurses per clinic was 1.97 (IQR 1.38, 3.29), AHPs per
on the number of currently active patients. Distances clinic was 0.60 (IQR 0.04, 1.27) and other employees per
to the nearest hospital and to Darwin or Alice Springs clinic was 2.20 (IQR 0.85, 3.85). The median number of
(whichever was closer) were measured using Google nurse, AHP and other position types shown on 2015–
Maps straight line distances in kilometres. The 2015– 2016 organisational charts for these remote health services
2016 NTG DoH Top End Health Service and Central was 2.0 (IQR 2.0, 4.0), 1.0 (IQR 1.0, 4.0), and 2.5 (IQR 1.7,
Australia Health Service organisational charts were 4.6) respectively.
used to determine the number of FTE nurse and AHP The median straight line distance to the nearest hospital
positions at each remote health service. Summary sta- of these health services was 205 km (IQR 143, 248), while
tistics were reported as means with 95% confidence in- the median distance to Alice Springs or Darwin was
tervals or medians with interquartile ranges (IQR). 247 km (IQR 180, 390). The median population catchment
Pearson correlations were used with a significance test for the health services was 460 (IQR 236, 798).
Russell et al. Human Resources for Health (2017) 15:52 Page 5 of 12

Overall, 347 nurse and AHP individuals ceased provid- are less extreme when defined as either leaving remote
ing care in any remote health service (secondary turnover services, they are nevertheless highly unstable and far
profile) during the 2013–2015 period. This was an annual higher than what has been reported in NT and in compar-
turnover rate of 65.8% (95%CI 58.3, 74.1) for nurses and able contexts elsewhere [17, 18]. Sub-analysis of unpub-
AHPs combined, with 74.4% (95%CI 65.3, 84.7) for nurses lished NTG DoH data by individual professional groups
and 37.8% (95%CI 27.7, 51.6) for AHPs. shows that the organisational turnover from NTG DoH
The mean overall annual turnover rate at the clinic was 34% for nurses (32%–36%), 22% for AHPs (13%–30%)
level (primary turnover profile) for nurses and AHPs and 33% for all staff (32%–34%) between 2013 and 2015.
combined was estimated as 128% (95% CI 114, 144) This is consistent with our study which found that turn-
(Table 1). Annual stability rates averaged 55.3% (95% CI over rates were significantly higher for nurses compared
49.6, 61.6) for nurses and AHPs. The mean probability to AHPs. While our study was not designed to explore the
that nurses and AHPs stay at least 12 months is 0.20 underlying reasons for differences in turnover of nurses
(95% CI 0.16, 0.24), and within 0.34 (95% CI 0.27, 0.42) compared to AHPs, it is likely that an important effect re-
years of commencement, half the nurses and AHPs lates to many AHPs working in communities located on
have left. Point estimates of the 12-month survival their traditional land and in which they have extended
probability and median survival of AHPs were longer family. The study also confirms anecdotal reports that
than for nurses, although this difference was not statisti- agency nurses provide a substantial proportion of primary
cally significant (Tables 2 and 3). Average annual stability care in NT remote communities, without whose services
rates, however, were significantly higher for AHPs at there would undoubtedly be significant gaps in the avail-
76.3% (95% CI 63.5, 91.7) compared with 48.3% (95% CI ability of primary care services.
42.3, 55.2) for nurses. These findings have important implications for remote
Remote NT health services had high levels of agency- health services more generally, their patients and for
employed nurse use. Of the 51 health services that had policymakers. Firstly, very high turnover rates mean that
nurse positions, agency-employed nurses provided a me- health services in remote communities will need to in-
dian of 0.40 FTE per clinic (IQR 0.26, 0.58). On average, vest considerable resources to adequately prepare and
for every 1.0 FTE nurse position on organisational charts, orient new staff to the health service and community
there were 0.15 FTE (95% CI 0.10, 0.23) agency-employed [22]. Of course, there is a trade-off between investing
nurses. resources in orienting new staff and investing those
Annual turnover rates for nurses and AHPs combined resources directly in providing health services for the
were significantly correlated with other workforce met- community, particularly in an under-resourced environ-
rics, though not with agency-employed nurse FTE ratios ment. Clearly, the shorter the period of time new staff
(Table 4). intend to stay in a remote clinic, the more the balance
Correlations between different workforce metrics and shifts towards investing directly in providing health ser-
health service characteristics were mostly statistically vices and away from providing extensive orientation,
non-significant. Annual turnover rates, however, had a since the return on investment in orienting new staff will
weak negative correlation with an indicator of staff supply be small for the most short-term staff. Nevertheless, it
(average number of nurses and AHPs actually working). remains important that new staff are trained to provide
The experienced nurse or AHP indicator had positive cor- appropriate and culturally safe care.
relations of moderate strength with several indicators of Secondly, lower turnover and higher stability rates
health service size (community population size, average amongst AHPs compared to nurses suggest that remote
number of nurses and AHPs actually working and average workforce stability may be better supported by greater
number of FTE nurse and AHP positions on 2015 organ- career development and employment opportunities for
isational chart) (Table 5). local Aboriginal community members to become AHPs
and nurses. Increased employment of Aboriginal local
Discussion community members in a range of other positions, includ-
This landmark study is one of very few studies that meas- ing administrative (community workers, alcohol and other
ure turnover and retention from the perspective of a par- drug workers, etc.) and logistic support roles (drivers,
ticular health service. Taking this perspective is important cleaners, gardeners) may similarly help provide improved
because it reflects what the experience of health profes- overall health workforce stability and simultaneously im-
sional continuity of care might be like from the viewpoint prove accessibility, quality of care and cultural appropri-
of consumers in a remote community. The study reveals ateness of health care [27].
extremely high annual turnover and poor retention for Thirdly, high staff turnover and low stability rates are
nurses and AHPs in remote NT health services. While also likely to result in the already limited funding available
combined nurse and AHP turnover and retention rates for remote health services being substantially less than
Table 1 Summary of nurse and Aboriginal health practitioner workforce metrics, by community population size and type
Russell et al. Human Resources for Health (2017) 15:52

Population size category of remote Number of clinics Mean annual nurse and Mean nurse and AHP Mean number of pay periods Mean survival probability for Average survival time
community AHP turnover (%) (95% CI) annual stability (%) per year with an experienced nurses and AHPs after 12 for nurses and AHPs
(95% CI) nurse or AHP (95% CI)a months (95% CI) in years (95% CI)
<200 predominantly Aboriginal 7 120 (68, 209) 61.9 (37.6, 100.0) 9.7 (8.4, 11.1) 0.12 (0.03, 0.29) 0.38 (0.08, 0.88)
200–349 predominantly Aboriginal 12 172 (128, 232) 54.9 (40.5, 74.4) 17.0 (15.7, 18.4) 0.10 (0.04, 0.18) 0.19 (0.11, 0.31)
350–799 predominantly Aboriginal 16 149 (120, 185) 53.6 (43.5, 66.0) 18.6 (17.4, 19.8) 0.17 (0.11, 0.24) 0.23 (0.15, 0.31)
≥800 predominantly Aboriginal 11 116 (97, 138) 54.5 (46.4, 64.1) 23.3 (21.7, 25.0) 0.25 (0.19, 0.31) 0.46 (0.38, 0.54)
Predominantly non-Aboriginal 7 82 (56, 120) 60.8 (44.0, 83.9) 17.7 (16.0, 19.6) 0.24 (0.12, 0.38) 0.31 (0.11, 0.69)
Total 53 128 (114, 144) 55.3 (49.6, 61.6) 17.9 (17.3, 18.6) 0.20 (0.16, 0.24) 0.34 (0.27, 0.42)
Turnover and stability calculations done at an aggregate level for each category of population size
AHP Aboriginal health practitioners, CI confidence interval
a
Poisson distribution assumed
Page 6 of 12
Table 2 Summary of nurse workforce metrics, by community population size and type
Population size category of Number of Mean annual nurse Mean nurse annual Mean number of pay periods Mean survival probabilityb Average survival time Mean agency-employed
a
remote community clinics turnover (%) (95% CI) stability (%) (95% CI) per year with a nurse who has for nurses after 12 months for nurses in years nurse FTE ratio
remained for 2 or more years (95%CI) (95% CI)
Russell et al. Human Resources for Health (2017) 15:52

c
(95% CI)
<200 predominantly Aboriginal 5 156 (76, 318) 57.9 (27.1, 100.0) 2.4 (1.8, 3.2) 0.14 (0.02, 0.36) 0.69 (0.04, 0.92) 0.04 (0.00, 1.68)
200–349 predominantly 12 202 (145, 281) 45.1 (30.9, 65.6) 8.3 (7.4, 9.3) 0.11 (0.05, 0.19) 0.19 (0.08, 0.31) 0.14 (0.04, 0.43)
Aboriginal
350–799 predominantly 16 175 (136, 224) 47.2 (36.2, 61.5) 14.6 (13.5, 15.7) 0.13 (0.07, 0.19) 0.19 (0.15, 0.27) 0.16 (0.07, 0.35)
Aboriginal
≥800 predominantly 11 130 (107, 158) 46.1 (37.5, 56.5) 18.9 (17.5, 20.5) 0.24 (0.18, 0.31) 0.42 (0.34, 0.54) 0.17 (0.10, 0.31)
Aboriginal
predominantly non-Aboriginal 7 75 (50, 112) 59.1 (42.1, 83.1) 16.1 (14.4, 17.9) 0.27 (0.13, 0.42) 0.50 (0.15, 0.77) 0.10 (0.02, 0.46)
Total for nurses 51 148 (126, 163) 48.3 (42.3, 55.2) 12.6 (12.1, 13.2) 0.19 (0.15, 0.23) 0.31 (0.23, 0.38) 0.15 (0.10, 0.23)
Turnover and stability calculations done at an aggregate level for each category of population size
CI confidence interval, FTE full-time equivalent
a
Annual stability rates were not estimated for two health services (both <200 population size category)
b
Survival probability after 12 months was inestimable for one health service (<200 population size category)
c
Poisson distribution assumed
Page 7 of 12
Table 3 Summary of Aboriginal health practitioner workforce metrics, by community population size and type
Population size category of remote community Number of clinics Mean annual AHP Mean AHP annual Mean number of pay periods Mean survival probability Average survivalb time
Russell et al. Human Resources for Health (2017) 15:52

turnover (%) (95% CI) stability (%) (95% CI) per year with an AHP who has for AHPs after 12 months for AHPs in years (95% CI)
remained for 2 or more years (95% CI)
(95% CI)a
<200 predominantly Aboriginal 6 75.7 (29.7, 193.4) 75.0 (37.9, 100.0) 8.0 (6.8, 9.3) 0.09 (0.00, 0.35) 0.31 (0.08, 0.88)
200–349 predominantly Aboriginal 7 66.5 (29.4, 150.4) 87.1 (51.4, 100.0) 9.7 (8.7, 10.8) 0.00 (−, −) 0.23 (0.15, −)
350–799 predominantly Aboriginal 14 93.1 (58.6, 147.9) 73.6 (52.9, 100.0) 8.8 (8.0, 9.7) 0.45 (0.23, 0.64) 0.77 (0.23, 1.92)
≥800 predominantly Aboriginal 11 65.8 (42.6, 101.7) 75.6 (57.7, 99.1) 17.8 (16.3, 19.3) 0.30 (0.12, 0.50) 0.46 (0.15, 1.03)
Predominantly non-Aboriginal 4 169.0 (53.0, 539.1) 77.8 (28.4, 100.0) 3.7 (2.9, 4.6) 0.00 (−, −) 0.04 (0.04, −)
Total for AHPs 42 79.4 (60.5, 104.2) 76.3 (63.5, 91.7) 10.1 (9.6, 10.6) 0.27 (0.16, 0.39) 0.46 (0.23, 0.84)
Turnover and stability calculations done at an aggregate level for each category of population size
AHP Aboriginal health practitioners, CI confidence interval, FTE full-time equivalent
a
Poisson distribution assumed (and the superscript in the cell with mean number of pay periods)
b
Median survival inestimable for seven remote health services (two services in <200, three services in 200–349, one in 350–799, one in ≥ 800 population size categories)
Page 8 of 12
Table 4 Correlations between different workforce metrics for nurses and Aboriginal health practitioners, 2013–2015 data
Russell et al. Human Resources for Health (2017) 15:52

Annual nurse and Nurse and AHP Number of pay periods per year Survival probability Median survival Agency-employed
AHP turnover annual stability with a nurse or AHP who has after 12 months time nurse FTE ratio
remained for 2 or more years
Annual nurse and AHP turnover 1
Nurse and AHP annual stability − 0.3103* 1
Number of pay periods per year with a nurse or AHP who has − 0.3542** 0.5427** 1
remained for 2 or more years
Survival probability after 12 months − 0.5665** 0.1286 − 0.0559 1
Median survival time − 0.4595** − 0.1432 − 0.1235 0.6732** 1
Agency-employed nurse FTE ratio 0.2368 − 0.1276 0.0593 0.0598 − 0.2146 1
AHP Aboriginal health practitioners, FTE full-time equivalent
*P < 0.05, **P < 0.01
Page 9 of 12
Russell et al. Human Resources for Health (2017) 15:52 Page 10 of 12

Table 5 Correlations between workforce metrics for nurses and Aboriginal health practitioners 2013–2015 and health service
characteristics
Health service characteristic Annual turnover Stability after 2 years + experienced 12 month survival Median Agency-employed
12 months nurse or AHP probability survival nurse FTE ratio
Distance to the nearest hospital 0.1071 − 0.1872 − 0.1074 − 0.0720 − 0.0499 0.2182
Distance to Darwin or Alice Springs − 0.0146 − 0.2303 − 0.0156 0.0073 − 0.0419 0.2378
Community population size − 0.2183 − 0.0587 0.3846** 0.1080 0.1255 0.1757
Average number of nurses and − 0.2929* − 0.0270 0.5163** 0.0419 0.1232 0.0900
AHPs actually working
Average number of FTE nurse and AHP − 0.1799 − 0.1856 0.4076** 0.0167 0.1575 0.0831
positions on 2015 organisational chart
*P < 0.05; **P < 0.01
AHP Aboriginal health practitioners, FTE full-time equivalent

would otherwise be available because of the excess costs services which may have only one or two key nurses or
of recruitment, agency fees and transport, orientation and AHPs, the health service workforce can be highly un-
induction, housing and other higher costs for new staff stable and switch between periods of relative staffing sta-
and for agency staff [27]. Extremely high turnover rates bility and periods of high use of short-term staff. This
may also increase risks to staff safety as constantly chan- study developed a new metric to indicate the number of
ging staff may be associated with decreased awareness of pay periods for which a health service had at least one
occupational health safety hazards [28]. Vacancies associ- nurse or AHP who had at least 2 years’ clinical experi-
ated with staff turnover may also exacerbate safety risks ence in that community. It is intended that each of the
for remaining resident staff, as fewer than the optimal turnover or retention metrics used in this study, includ-
numbers of staff are available to deliver services. High ing the new metric, will be tested to assess their useful-
turnover and poor retention, together with high use of ness in predicting quality of primary care and potentially
short-term agency staff, are also likely to limit the ability avoidable hospitalisations. While the usefulness of this
of the health service to provide high-quality care and par- metric is yet to be tested in other comparable contexts,
ticipate meaningfully in continuous quality improvement our research nevertheless corroborates the use of multiple
activities [27]. Lack of stable resident primary care pro- well-established workforce metrics, including annual turn-
viders is also likely to compromise the effectiveness of vis- over rates, annual stability rates and survival probabilities,
iting specialist and allied health services, since primary to provide a comprehensive picture of patterns of work-
care providers in remote communities have a key role to force turnover and retention in remote health services.
identify and prioritise who needs to be seen for what and We found moderate to strongly significant correlations
subsequently to implement and monitor any required between annual turnover rates and other workforce met-
follow-up. rics suggested by Russell et al. for use in rural health ser-
For patients, high use of agency staff by health ser- vices [9]. Most workforce metrics, with the exception of
vices, in conjunction with low retention and high turn- experienced staff, were not significantly correlated with
over of NTG DoH-employed staff, results in relational health service characteristics. This perhaps reflects that
discontinuity with their primary care providers. This is the 53 health services in the study were all at the extreme
particularly important for Aboriginal patients, especially end of the spectrum of staff turnover and retention expe-
those with more serious or chronic health problems, as rienced in Australian health services and therefore unable
it means that the time needed for them to feel culturally to be differentiated according to community population
safe and begin trusting their primary care providers is size or distances to the nearest hospitals or to Darwin/
not available. As a result, patients may be less likely to Alice Springs.
access the care they need in a timely way [29]. Further, it Given the substantial policy significance of these
is more likely that they will experience health encounters research findings, it is important to acknowledge sev-
where the primary care provider is not adequately pre- eral limitations of the data and analysis. Firstly, data on
pared for working in a complex cross-cultural environ- agency-employed nurses were not available at an individ-
ment. Ultimately, the quality of care that they receive is ual level, could not be integrated with payroll data and
likely to be lower and their health outcomes poorer. were not recorded in a sufficiently accurate or detailed
Some of the workforce metrics that were calculated in way to enable complete capture of all agency nurses at
this analysis have not previously been reported in the each remote health service. The turnover rates reported
published, peer-reviewed literature to describe workforce are therefore underestimates, since they exclude agency
turnover and retention patterns. In small, remote health nurses paid directly by agencies. Similarly, stability rates
Russell et al. Human Resources for Health (2017) 15:52 Page 11 of 12

may represent overestimates of the overall patterns. competently managed and clinical protocol driven. It is
Anecdotal information from health service providers also crucial that health workforce policies are developed
confirm these assertions. The financial expenditure for that effectively stabilise the remote primary care workforce
labour hire costs, used to derive agency-employed and optimise workforce turnover because the benefits of
nurse FTE, however, are considered to be accurate and continuing heavy reliance on short-term nursing staff are
reliable and have been used in our research to provide offset, and at times entirely negated by, substantial down-
a comprehensive overall picture of staffing patterns in sides. Stabilising the remote workforce may require
remote communities. different workforce models to be utilised. For example, in-
Secondly, it was not possible to allocate staff on the dividuals may work 1 month on, 1 month off in remote
NTG DoH payroll working in a supernumerary capacity communities in shared positions. Other possible solutions
to specific health services if the cost centre covered include emulating medical workforce training strategies
multiple remote services. However, these comprised with preferential selection of rural or remote students into
only a small proportion of nursing and AHP staff. Simi- nursing and AHP training courses, providing vocational
larly, we were unable to allocate all of the aggregated training based in rural and remote settings, supporting
agency-employed nurse FTE data to specific remote students with remote scholarships and providing recruit-
health services, as some of the cost centres covered ment and retention incentives for working in remote loca-
multiple remote services. Further, our definition of tions once students graduate. Another strategy may be to
agency-employed nurse excludes those agency nurses eliminate barriers that remote dwelling Aboriginal Austra-
paid directly by NTG DoH, and our analysis has not lians face when entering and remaining in the health
attempted to specifically identify agency nurses on workforce. These barriers include English literacy and
NTG DoH casual or temporary contracts. Our overall numeracy levels, adequacy of remuneration for AHPs, and
reported agency-employed nurse FTE ratio of 0.15 is employment conditions such as lack of subsidised hous-
therefore likely to be a substantial underestimate of ing, each of which may act as a deterrent. In the face of
total NTG DoH use of agency nurses. One recent re- geographical maldistribution of doctors, community-
port quotes a figure of 42% of remote nursing positions based remote nurse practitioners may help to stabilise the
filled by agency-employed and NTG DoH-employed workforce and improve access.
agency nurses [22]. Anecdotal evidence from nurse co-
ordinators also indicates that between a third and a half Conclusions
of remote NT nursing positions are currently filled by This research provides rigorous empirical evidence that in
agency nurses. remote NT communities with NTG DoH health services,
Thirdly, in focussing our analysis on nurses and turnover of nurses and Aboriginal health practitioners is
AHPs, we do not capture all clinical and non-clinical extremely high, stability rates are low and substantial use
staff working at remote health services. Remote health is made of agency nurse services. These staffing patterns
services function with the support of resident adminis- are almost certainly contributing to sub-optimal continu-
trative staff, Aboriginal community workers, alcohol and ity of care, compromised health outcomes, poorer levels
other drugs workers, Aboriginal mental health workers, of staff safety and higher costs. To effectively address these
cleaners, drivers, gardeners, receptionists, community li- deficiencies, it is imperative that Territory and Federal
aison officers, healthy lifestyle educators and so on. A Governments invest in implementing, adequately resour-
small number of remote communities also have resident cing and evaluating staffing models which effectively sta-
doctors for whom workforce mobility information was not bilise the remote primary care workforce as a matter of
available. Further, remote health clinics are often sup- priority. The results are also important for quantifying
ported by visiting outreach workers providing medical, workforce patterns in a rural or remote area, a subject for
nursing and allied health services across a range of areas which there has been substantial national and inter-
including public health, continuous quality improvement, national interest but limited research.
health promotion, preventable chronic disease coordin-
Abbreviations
ation and across various health care specialties. This ana-
95%CI: 95% confidence interval; AHP: Aboriginal health practitioner;
lysis, therefore, while focused on the nurse and AHP AHW: Aboriginal health worker; DoH: Department of Health; FTE: Full-time
workforce who are resident in communities, does not cap- equivalent; GAS: Government accounting system; IQR: Interquartile range;
NT: Northern Territory; NTER: Northern Territory emergency response;
ture the entire spectrum of health providers working in
NTG: Northern Territory Government; PIPS: Personnel information and
these communities. payroll systems
Nonetheless, these limitations notwithstanding, the im-
plications of the findings of this research for policymakers Acknowledgements
We wish to thank Steve Kyriacou, who provided technical assistance in
are profound. In the face of such high turnover, we need accessing and interpreting departmental personnel and financial data and
robust health service models that are adequately funded, Professor David Lyle for feedback we received on our manuscript during its
Russell et al. Human Resources for Health (2017) 15:52 Page 12 of 12

drafting. We would also like to acknowledge and sincerely thank all staff 10. Waldman JD. Change your metrics: if you get what you measure, then
working in the 53 remote NT communities, particularly the clinic managers. measure what you want—retention. J Med Pract Manage. 2006;22:13–9.
11. Waldman JD. Measuring retention rather than turnover: a different and
Funding complementary HR calculus. Hum Resour Plan. 2004;27:6–9.
This project was supported under Australian Research Council’s Discovery 12. Dolea C, Stormont L, Braichet JM. Evaluated strategies to increase attraction
Projects funding scheme (project number DP150102227). and retention of health workers in remote and rural areas. Bull World Health
Organ. 2010;88:379–85.
Availability of data and materials 13. Pathman DE, Konrad TR, Ricketts TC. The comparative retention of National
The datasets generated and analysed during the current study are not Health Service Corps and other rural physicians. Results of a 9-year follow-up
publicly available due to identifiability of remote primary care providers and study. JAMA: Journal of the American Medical Association. 1992;268:1552–8.
the need to protect their privacy. 14. Jones CB. The costs of nurse turnover, part 2: application of the nursing
turnover cost calculation methodology. J Nurs Adm. 2005;35:41–9.
Authors’ contributions 15. Nursing Solutions Inc. National healthcare retention & RN staffing report,
This research was conceived by JW, JH, SG, YZ and MJ. The study was vol. 2016. East Petersburg, PA: NSI; 2016.
planned and coordinated by JW, SG, DR, MR, YZ, MJ and JH. Contributors to 16. Kosel KC, Olivo T: The business case for workforce stability. Irving, TX; VHA
the analysis and interpretation were DR, YZ, MJ, SG, MR and JW. All authors Inc.; 2002.
helped draft the manuscript and read and approved the final manuscript. 17. Garnett ST, Coe K, Golebiowska K, Walsh H, Zander KK, Guthridge S, Li S,
Malyon R. Attracting and keeping nursing professionals in an environment
of chronic labour shortage: a study of mobility among nurses and midwives
Ethics approval and consent to participate
in the Northern Territory of Australia. Darwin: Charles Darwin University
Ethics approval was received from the Human Research Ethics Committee of
Press; 2008.
the Northern Territory Department of Health and Menzies School of Health
18. Health Q. Ministerial taskforce: nursing recruitment and retention. Final
Research (2015-2363).
report. Queensland Health: Brisbane; 1999.
19. Boffa JD, Bell AI, Davies TE, Paterson J, Cooper DE. The Aboriginal Medical
Consent for publication Services Alliance Northern Territory: engaging with the intervention to
Not applicable. improve primary health care. Med J Aust. 2007;187:617–8.
20. Studdert L. Remote Area Health Corps: nurses making a contribution to
Competing interests primary health services in the NT. Australas Emerg Nurs J. 2010;13:142.
The authors declare that they have no competing interests. 21. Busbridge MJ, Smith A. Fly in/fly out health workers: a barrier to quality in
health care. Rural Remote Health. 2015;15:3339.
22. Northern Territory Government Department of Health. Remote area nurse
Publisher’s Note safety: on-call after hours security. Darwin: NT Department of Health; 2016.
Springer Nature remains neutral with regard to jurisdictional claims in
23. Health Workforce Australia. Aboriginal and Torres Strait Islander Health
published maps and institutional affiliations.
Worker project. Adelaide: HWA; 2011.
24. Australian Institute of Health and Welfare. Aboriginal and Torres Strait
Author details
1 Islander health practitioner workforce 2014. Canberra: AIHW; 2016.
Monash Rural Health, Monash University, PO Box 666, Bendigo, Victoria
25. Australian Bureau of Statistics. Estimates of Aboriginal and Torres Strait
3552, Australia. 2Department of Health, PO Box 40596, Darwin, NT 0800,
Islander Australians, Jun 2011. Cat No. 3238.0.55.001. Canberra: ABS; 2013.
Australia. 3Faculty of Human Sciences, Macquarie University, North Ryde,
26. Northern Territory Government Department of Health. Indicator definition:
NSW 2109, Australia. 4Monash Rural Health, Monash University, PO Box 91,
full time equivalents v1.0. Darwin: NTG Department of Health; 2015.
Strathdale, Victoria 3550, Australia. 5Flinders Northern Territory, School of
27. Parliament of Australia. Report on the inquiry into nursing. The patient
Medicine, Flinders University, PO Box U362, Casuarina, NT 0815, Australia.
profession: time for action. Canberra: Commonwealth of Australia; 2002.
28. CRANAplus. Remote health workforce safety and security report: literature
Received: 16 February 2017 Accepted: 7 August 2017
review, consultation and survey report. Cairns, Australia: CRANAplus; 2017.
29. Guthrie B, Saultz JW, Freeman GK, Haggerty JL. Continuity of care matters.
BMJ. 2008;337:a867.
References
1. World Health Organization. The World Health Report 2006: working
together for health. Geneva: WHO; 2006.
2. Sharma DC. India still struggles with rural doctor shortages. Lancet.
2015;386:2381–2.
3. Sommanustweechai A, Putthasri W, Nwe ML, Aung ST, Theint MM,
Tangcharoensathien V, Wynn SS. Community health worker in hard-to-
reach rural areas of Myanmar: filling primary health care service gaps.
Hum Resour Health. 2016;14:64.
4. Terschuren C, Mensing M, Mekel OCL. Is telemonitoring an option against
shortage of physicians in rural regions? attitude towards telemedical
devices in the North Rhine-Westphalian health survey, Germany. BMC
Health Serv Res. 2012;12:95. Submit your next manuscript to BioMed Central
5. Dove N. Can international medical graduates help solve Canada's shortage and we will help you at every step:
of rural physicians? Canadian Journal of Rural Medicine. 2009;14:120.
6. Vujicic M, Shengelia B, Alfano M, Thu HB. Physician shortages in rural • We accept pre-submission inquiries
Vietnam: using a labor market approach to inform policy. Soc Sci Med. • Our selector tool helps you to find the most relevant journal
2011;73:970–7.
• We provide round the clock customer support
7. Kippenbrock T, Lo W-J, Odell E, Buron B. The Southern states: NPs made an
impact in rural and healthcare shortage areas. Journal of the American • Convenient online submission
Association of Nurse Practitioners. 2015;27:707–13. • Thorough peer review
8. Mason J. Review of Australian government health workforce programs.
• Inclusion in PubMed and all major indexing services
Canberra: Australian Government Department of Health; 2013.
9. Russell DJ, Humphreys JS, Wakerman J. How best to measure health • Maximum visibility for your research
workforce turnover and retention: five key metrics. Aust Health Rev.
2012;36:290–5. Submit your manuscript at
[Link]/submit