Post-polio hospital admissions in Australia over a 10-year period: An observational study and analysis of trends by month, location, and comparable conditions

Abstract

Background

There is currently no precise estimate of post-polio conditions in Australia. This observational study aimed to provide a summary of hospitalisations over a 10-year period in Australia, with a specific focus on annual, monthly, and regional trends, as well as a comparison with four similar neurological conditions.

Methods

A retrospective cohort study of late effects of polio and post-polio syndrome from 2011 to 2021. Primary data were obtained via a data-on-request process through the Australian Institute of Health and Welfare in 2022. Analysis was carried out according to Welch Analyses of Variance with Games-Howell post-Hoc tests using GraphPad PRISM and Stata Version 17.0.

Results

There was a statistically significant decrease in the monthly hospitalisations over the 10-year period and months according to seasonal trends, and significant differences across geographical regions and regionality. Significant differences existed between the number of hospitalisations per 100 000 population across diagnostic codes (p < 0.01).

Conclusions

Annual trends were identified from 2010 to 2021 for post-polio hospitalisations in Australia. This research improves the precision of estimates for post-polio conditions in Australia, and provides helpful information on where people are hospitalized in Australia. These estimates are internationally comparable and can inform clinicians and health service managers worldwide.

communicable diseases, epidemiology, epidemiology

Introduction

Late Effects of Polio (LEoP) and Post-Polio Syndrome (PPS) are the terms predominantly used in recent research literature, and preferred within Australia, when referring to post-polio conditions.^1,2 LEoP captures the biomechanical sequelae of acute polio damage, while PPS reflects neurological deterioration and is a primary diagnosis based on criteria.² The cluster of post-polio symptoms, which emerge some 15 years or more after infection, is variable in its onset, severity and progression across individuals, and includes fatigue, progressive weakness, complex pain, sleep disturbance, breathing and swallowing difficulty, and cold intolerance.^2,3

Hospitalisation incidence for those with post-polio conditions are suspected to be above age norms but below those with more fragile or debilitating health conditions. The greatest risks to life lay in the high incidence of new respiratory problems (58% of community-living Australian survivors annually)⁴ and serious falls among polio survivors.⁵ Also, polio survivors in Denmark are reported to have a 1.2- to 1.3-fold increased risk of hospitalisation for pulmonary, heart, gastrointestinal disorders, or paralysis of different severities affecting one or more limb.⁶ Literature also suggests that polio survivors have a higher rate of comorbid conditions such as depression,^7,8 falls,⁷ dyslipidaemia,⁹ sleep apnoea, hypoventilation and restless legs syndrome¹⁰, and being overweight.¹¹

A 2012 systematic review of the worldwide prevalence of people with post-polio conditions highlighted significant variability between countries and the need for studies of national samples, particularly in high-income countries.¹² Australian prevalence data exists for other neuromuscular conditions, of which only some are similar to worldwide estimates^1,12–29 (see Table 1). In Australia, multiple sclerosis (MS)¹⁸ appears to have a comparable prevalence to, and Parkinson’s disease (PD)¹⁴ appears to have double the prevalence of post-polio conditions.

Table 1

Prevalences of neuromuscular conditions.

Neuromuscular condition	ICD-10-AM of conditions in this analysis	Australian prevalence, per 100 000	Worldwide prevalence, per 100 000
Parkinson’s disease (PD)	G20	294	106
Sequelae of poliomyelitis (LEoP), and Post-polio syndrome (PPS)	B91 and G14	160#	15 to 1773
Multiple sclerosis (MS)	G35	131	36
Non-traumatic spinal cord injury (NTSCI)		37	112 to 231
Charcot Marie Tooth disease (CMT)		40	10 to 82
Myasthenia gravis (MG)	G70.0	12	8
Motor neuron disease (MND)*	G12.2	9	3
Huntington’s disease (HD)		8	5
Inclusion body myositis (IBM)^		5	5
Spinal muscle atrophy (SMA)		3	2

^#Mean of several of Polio Australia’s internal estimates. Polio Australia is the peak representative body for people with post-polio conditions in Australia and is committed to standardizing quality polio information and service provision across Australia for polio survivors: https://www.poliohealth.org.au/mission/  ^*MND prevalence primarily being amyotrophic lateral sclerosis (ALS). ^IBM prevalence primarily being sporadic IBM.

The polio-affected population’s prevalence (diagnosed and undiagnosed) in Australia is estimated to number in the tens of thousands and includes those who had identifiable viral symptoms up to and including paralysis after exposure to the poliomyelitis virus.¹ Although imprecise, this estimate is based on numerous historical records, disease reports, and recent cohort surveys and is comparable to other first world countries who experienced polio epidemics.³⁰

Hospitalisation rates have been used to provide information on the prevalence of health conditions through hospitals’ coding records, and act as one of a range of sources in disease prevalence studies in Australia and elsewhere.^20,31,32 As the identification and assessment of those exposed to polio who have or will develop post-polio symptoms is challenging, their prevalence and recognition in health systems is vital to reduce unnecessary risks during hospital care.^12,33 Unpublished survey responses in 2020 from 180 polio survivors known to Polio Australia revealed 43% had used a hospital in the previous 10 years.⁵ Hence, obtaining hospitalisation data presents as a key strategy in identifying the burden of disease of post-polio conditions in Australia, which can be compared internationally. This analysis would further help improve the accuracy of existing Australian polio survivor prevalence estimates.

This research aimed to provide detail on the incidence of post-polio diagnosis codes in hospitalisations over a ten-year period (2011–2012 to 2020–2021) in Australia. The primary research questions were:

1. What annual hospitalisation volumes and trends exist for those with post-polio diagnoses?

2. Is there a peak month or season of hospitalisation for those with post-polio diagnoses?

3. What do analyses of state, and capital city versus regional areas show about hospitalisations?

4. What are the concurrent coding incidences of similar neurodegenerative conditions and how do they compare to post-polio incidences?

Additional demographics drawn from a public AIHW data set further characterize those hospitalized with a diagnosis of PPS, and those with the four neurodegenerative conditions of interest.

Methods

Data collection and handling

Private and public hospitals in Australia utilize the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) to record and classify episodes of care.³⁴ LEoP aligns with the ICD-10-AM code B91 ‘Sequalae of poliomyelitis’, and PPS is titled such and coded as G14.

The primary B91/G14 data for this analysis were obtained via a data-on-request process through the Australian Institute of Health and Welfare (AIHW) in 2022, utilizing funding provided to Polio Australia by the Australian Department of Health. The decade of financial years data received (2011–2021), was tabled and characterized hospitalisations (separations) during the period, as outlined in Table 2. The data were categorized according to the four research questions. These data were non-identifiable, and sourced from public and private hospitals operating in the states and mainland territories of Australia. The annual separation incidence of five neuromuscular conditions of interest over the same 10-year period was included in the data, four indicated by their ICD-10-AM code being listed in Table 1. The fifth condition (G93.3 Postviral fatigue syndrome) was not included due to coding and definition discordance in its relation to Myalgic Encephalitis and Chronic Fatigue Syndrome over the corresponding decade.^35,36

Table 2

Financial year hospitalisations with post-polio diagnostic codes G14 or B91 in Australia (all regions, private and public).

Australian financial year (July 1st to June 30th)	Requested AIHW data		Public AIHW data
	Number of post-polio hospitalisations (G14 and B91)	Percent hospitalized in greater capital city area facilities	Number with a G14 primary diagnosis
2011–12	757	61%	No data
2012–13	770	55%	No data
2013–14	756	60%	No data
2014–15	835	55%	No data
2015–16	863	60%	109
2016–17	934	64%	151
2017–18	888	63%	199
2018–19	818	63%	179
2019–20	751	70%	186
2020–21	679	66%	85

Public data on post-polio hospitalisations published by the AIHW within the same period were also obtained. This data, also non-identifiable, spanned six financial years from 2015–2016 to 2020–2021 and included categories of age, sex, separation occurrences, and separation days. This public data only included primary diagnoses—for post-polio conditions only code G14. Data for the four included neuromuscular conditions, as primary diagnoses, were drawn concurrently from this source.

Data analysis

Descriptive analysis was carried out using GraphPad PRISM v.9.5.1 (GraphPad Software, LLC, Boston MA, USA). Trends according to each of the variables were presented graphically displaying mean ± SD for hospitalisations according to year, month, state, and diagnostic code. Analyses of variance were carried out for each of the research questions according to a range of methods, dependent on the nature of the variables. Multiple assumptions were tested including (i) outliers (via Boxplots), (ii) normality of distribution of data (Shapiro–Wilk Test) and (iii) homogeneity of variances (Levene’s Test) in the data. In instances where the assumptions were upheld (i.e. there were not violations to these assumptions), a standard analysis of variance (ANOVA) was carried out with Tukey post-hoc tests. In instances where there were substantial (i.e. extreme) outliers (identified as more than 2 SD), any extreme outlier was removed from the dataset or a suitable alternative test (e.g. Kruskal–Wallis H Test) was run to take this into account. Where there were non-normal distributions of the variables, transformations were carried out and explored as to whether using a transformed variable was most appropriate. In instances of heterogeneity of variances, the Welch ANOVA with Games-Howell post-hoc tests were carried out. Post-hoc simple and complex contrast tests were also carried out, where feasible, providing an indication of where specifically the differences were between variables (e.g. specific months of the year). All analyses were conducted using Stata v 17.0 (StataCorp, Texas, USA).

Results

1. What annual hospitalisation volumes and trends exist for those with post-polio diagnoses?

As outlined in Fig. 1a and b, there was an upward trend in monthly overall hospitalisations for polio survivors from 2011–2012 to 2016–2017, with a subsequent downward trend to 2020–2021. Given violation of the assumption of homogeneity of variances F (9,110) = 2.93, p = 0.004, a Welch ANOVA found statistically significant differences between the mean number of monthly hospitalisations across the years (Welch’s F (9,44.63) = 2.52, p = 0.0198).

Figure 1

(a) Monthly hospitalisations per year for polio survivors in Australia; (b) Total number of hospitalisations per year for polio survivors in Australia.

A Games-Howell post-hoc test identified a statistically significant decrease in the mean number of monthly hospitalisations from 77.8 ± 16.26 in 2016–2017 to 56.6 ± 10.05 in 2020–2021, a decrease of 21.25 ± 5.52 [mean ± standard error], p = 0.029.

1. Is there a peak month or season of hospitalisation for those with post-polio diagnoses?

As outlined in Fig. 2a and b, there were three main trends (winter-summer, monthly and isolated months) across the annual cycle for the mean number of hospitalisations across the 10-year period. Hospitalisations increased from January to March, decreased to April prior to increasing to August, and then decreased to December.

Figure 2

(a) Average number of hospitalisations per year for each month, and (b) Number of hospitalisations (per year) each month for polio survivors across the 10-year period.

Hospitalisations were highest over the winter months (June = 71.2 ± 11.9, July = 73.5 ± 16.1 and August = 73.8 ± 19.9), however there were also high numbers of hospitalisation in March (72.2 ± 21.8) and May (70.4 ± 23.4). The lowest months for hospitalisation were summer months (January = 59.8 ± 11.8 and December = 59.4 ± 9.50).

The mean hospitalisations each month were not statistically significant, F(11, 108) = 1.10, p = 0.372, when compared across the 10 years. Tukey post hoc analysis revealed no statistically significant mean differences across months (p > 0.05). Post-hoc contrasts revealed no statistically significant differences between January and August [t (SE) = 1.92 (7.21), 95% CI -0.30 to 28.3, p = 0.055], February and July [0.71 (7.21), 95% CI -9.20 to 19.40, p = 0.481] or summer and winter [t (SE) = 0.94 (4.16), 95% CI -4.32 to 12.19, p = 0.357]. However, there was a statistically significant difference between December (59.40 ± 9.50) and August (73.8 ± 19.89), [t (SE) = 2.00 (7.21), 95% CI 0.104 to 28.70, p = 0.048] and January/December and July/August [t (SE) = 2.75 (5.10), 95% CI 3.94 to 24.16, p = 0.007].

1. What do analyses of state, and capital city versus regional areas show about hospitalisations?

There were disparate hospitalisation trends by state: some states fluctuated (e.g. NSW), some decreased (TAS, VIC and WA), and some were sporadic (e.g. ACT, NT, QLD and SA). Figure 3 shows the differences between geographical region (capital cities versus not capital city areas) in states with the greatest raw hospitalisations, respectively. Further, Fig. S1 in the supplementary material outlines the differences in states with the least raw hospitalisations.

Figure 3

Comparison of trends between capital cities and regional locations, by the states with the greatest hospitalisations.

The one-way ANOVA carried out determined the average number of hospitalisations per month was different according to different states in Australia when normalized by population. Given violation of the assumption of normality and homogeneity of variances F (7,148) = 10.69, p < 0.01, a Welch ANOVA found statistically significant differences between the mean number of hospitalisations across states (Welch’s F(7, 61.48) = 12.119, p < 0.01).

For geographical region there was also a statistically significant difference between location (capital cities and non-capital cities) Welch’s F(1, 91.10) = 16.140, p < 0.01. Further, a Kruskal-Wallis H found differences in hospitalisations between two groups based on geographical location, with statistically significantly different median ranks of hospitalisations between groups, H(1) = 9.157, P = .0025 (capital cities N = 76, median = 53, non-capital cities N = 80, median = 15). As outlined in Table 3, Games-Howell post-hoc tests identified statistically significant differences between a range of states.

1. What are the concurrent coding incidences of similar neurodegenerative conditions and how do they compare to post-polio incidences?

Table 3

Comparisons between states for mean incidence per 100 000 across the 10-year time period; those with significant differences only.

Comparison	Means	Mean Difference [SE]	Games and Howe (T-Score)	95% CIs	p-value
SA and VIC	3.70 ± 3.35 and 0.83 ± 0.39	2.86 [0.75]	3.79	0.31 to 5.41	0.021
SA and NT	3.70 ± 3.35 and 1.03 ± 1.18	2.66 [0.80]	3.34	0.23 to 5.30	0.047
NSW and VIC	2.13 ± 0.68 and 0.83 ± 0.39	1.30 [0.18]	7.41	0.73 to 1.87	<0.01
TAS and VIC	1.94 ± 1.14 and 0.83 ± 0.39	1.11 [0.27]	4.04	0.19 to 2.03	0.011
NSW and WA	2.13 ± 0.68 and 1.03 ± 1.18	1.10 [0.31]	5.55	0.73 to 1.87	<0.01
NSW and NT	2.13 ± 0.68 and 1.03 ± 1.18	1.10 [0.31]	3.54	0.09 to 2.12	0.026
QLD and VIC	1.55 ± 0.35 and 0.83 ± 0.39	0.72 [0.12]	6.06	0.34 to 1.09	<0.01
NSW and QLD	2.13 ± 0.68 and 1.55 ± 0.35	0.59 [0.17]	3.41	0.03 to 1.15	0.036
QLD and WA	1.55 ± 0.35 and 1.03 ± 1.18	0.52 [0.15]	3.44	0.30 to 1.01	0.031

Ten-year annual hospitalisations for each of the four neurodegenerative conditions of interest, versus their respective population prevalence estimates are provided in Fig. 4. Raw values for these conditions are outlined in Supplementary Fig. 2. The proportions of hospitalisations to prevalence were relatively stable over the period for PD, MS and MND; having proportions of 0.02 for LEoP/PPS, 0.4 for PD, 1.0 for MS, and 1.5 for those with MND (i.e. expect an Australian with MS to have about one annual hospitalisation). The unique condition observed in this period was MG, in which case the proportion increased from 2.2 to 4.4 average hospitalisations per person with MG.

Figure 4

Standardized hospitalisations (hospitalisations/prevalence) over the 10-year period by diagnostic group. LEOP: Late effects of polio (G14), MG: Myasthenia gravis (G70), MND: Motor neurone disease (G12,2), MS: Multiple sclerosis (G35), PD: Parkinson’s disease (G20).

Figure 5 shows the number of hospitalisations per condition standardized for the Australian population over the relevant years, per 100 000 (i.e. a hospital serving 100 000 residents should expect these annual admissions). The order of expected admissions was MS, PD, MG MND, and LEoP/PPS (mean/median 3.4 per 100 000). Given violation of the assumption of homogeneity of variances F (4,45) = 8.87, p < 0.001, a Welch ANOVA was carried out that found statistically significant differences between the mean number of hospitalisations across the diagnostic groups (Welch’s F (4,19.44) = 907.108, p < 0.001). Based on public AIHW data, table 4 outlines the average annual hospitalisations and the days per hospitalisation based on similar neurodegenerative conditions. On average, there were 3.5 days per hospitalisation for those with post-polio syndrome. Table 5 provides age group deciles, according to the total hospitalisations for these similar neurodegenerative conditions from 2015 to 2021.

Figure 5

Standardized (per 100 000 population) hospitalisations over the 10-year period by diagnostic group. MG: Myasthenia gravis (G70), MND: Motor neurone disease (G12,2), MS: Multiple sclerosis (G35), PD: Parkinson’s Disease (G20).

Table 4

Public AIHW data (2015 to 2021) on comparable neurodegenerative conditions’ hospitalisations, ^*broad ranges.

Neuromuscular condition	Average annual hospitalisations	Average of hospitalisations per 100 000 Australians	Average annual total days in hospital	Average days per hospitalisation	Percent females of those hospitalized	Percent aged 65+ of those hospitalized
MND	1964	13.9	15 905	8.1	40–47	52–60
PD	13,243	119.5	70 933	5.4	34–36	84–88
PPS	152	3.4	530	3.5	64–77*	52–83*
MS	30,102	129.5	46 962	1.6	74–76	2–6
MG	11,210	42.1	16 042	1.4	57–61	47–52

Table 5

Public AIHW data (2015 to 2021) on comparable neurodegenerative conditions’ hospitalisations, grouped according to age group decile.

Age Group (Decile)	MND	PD	PPS	MS	MG
0 to 9	29	0	0	5	312
10 to 19	12	25	0	2273	908
20 to 29	91	16	2	21 842	3068
30 to 39	298	77	11	47 789	5269
40 to 49	819	763	43	53 141	7755
50 to 59	2394	3982	104	36 644	10,079
60 to 69	3368	16,626	274	14 306	13,982
70 to 79	3452	35,284	304	4098	16,579
80+	1322	22,684	171	511	9365

Discussion

Main finding of this study

Annual trends of post-polio hospitalisations in Australia from 2010–2011 to 2020–2021 were low within the estimated post-polio population, low within the Australian population, and the lowest amongst four additional neurodegenerative conditions compared within the same period.

What is already known on this topic

This is the first study internationally to examine post-polio hospitalisations incidences and characteristics in a single country over a ten-year period. Hospitalisation incidences have been used in research elsewhere to inform health conditions’ prevalences, trends and health policy, and to estimate health system use and costs of those with specific health conditions.^31,37

What this study adds

Raw hospitalisation numbers for LEoP/PPS over the decade of data revealed an upward then downward trend nationally, but were consistent when considered as standardized proportion of the population. Significant differences were observed between warmest and coldest months, and across states and capitals versus regional areas. Annual data on four neuromotor conditions within the period highlighted differences in their features and trajectories.

Annual Trends

LEoP/PPS hospitalisation incidences demonstrated an inverse-U trend with a peak around the year 2016–2017. Our analysis demonstrated a statistically significant difference between 2016–2017 and 2020–2021. Data demonstrated a consistent standardized hospitalisation rate of 2 per 100, this being below Polio Australia’s estimate of 8 per 100 derived from a convenience sample in which SA respondents were over represented.⁵

The rise of hospitalisations may reflect a gradual increased use of hospitals by this population as they age and as their post-polio conditions emerge or progress. This decade aligns with the development and rollout of the National Disability Insurance Scheme.³⁸ For those under age 65 who were eligible to access the NDIS, increased community health support may have contributed to the reduction in hospitalisations in the data’s latter years. Additionally, during the decline period, Polio Australia’s professional education program for multidisciplinary clinicians was active. National mitigation measures and hospital bed shortages during the COVID-19 Pandemic were a likely basis for low 2020–2021 hospitalisations. Due to the pandemic it has been well documented that during this period there was a lower risk of general infection,³⁹ which may have also decreased the hospitalisations during this period.

Monthly trends

Although there were no significant differences when considering monthly trends across the period, contrasts identified significant differences when evaluating December and January against July and August, demonstrating a seasonal early summer and late winter trend. This is consistent with the literature that outlines a relationship between the cold winter months and an increase in infections.⁴⁰

Several chronic conditions have seasonal variations in Australian hospitalisations.⁴¹ Cold weather can limit some polio survivors’ activity and participation.⁴² One symptom of post-polio is cold intolerance, with an incidence of 25% to 56%⁴³ and which worsens other LEoP symptoms.⁴⁴ Local variables such as hospital staffing and the environment can influence hospital admission timing.⁴¹

In cooler seasons, those with post-polio conditions can experience symptoms causing susceptibility to falls. Without data on the reasons or qualities of hospitalisations we cannot exclude other causes. Nevertheless, these insights are useful to identify times of pronounced post-polio hospitalisations.

State trends

The number of hospitalisations by state averaged over the year were considerably different across months. NSW had substantially greater raw numbers, so incidence per 100 000 population provided a more realistic representation of differences. SA had the most and VIC had the least hospitalisations per capita during this period.

Several factors may account for the higher incidence of SA hospitalisations. The proportions of those managed in hospitals may indicate (inversely) the capacity of community health, or higher rates of comorbidities. Higher hospitalisations in the polio-affected in SA may relate to their older mean age and higher likelihood to have had LEoP onset. VIC has a low hospitalisation rate, both raw and per-capita. This may be explained by their gold standard post-polio service infrastructure. This state has a designated post-polio clinic which also provides a mobile clinic that travels to six regional cities annually, and the state has two active post-polio support organisations. These factors provide those affected by polio in VIC with clinical and peer support to aid LEoP/PPS management. In addition, other specialist services are positioned centrally and close to the state’s population.

Regional trends

Hospitalisations within each of the largest five states reflect total capital to regional population distributions.⁴⁵ Data suggests those affected by polio are spread relatively proportionally between capital and regional areas. QLD and NSW have an almost even split between capital and regional LEoP/PPS hospitalisations; the QLD total population is similarly split. NSW and VIC hospitalisations appear distorted towards regional services when contrasting to population. This might be explained by the quality of or access to regional services, or by health disparity in regional areas. VIC, WA and SA have hospitalisations and state populations weighted to their capitals, SA being the most distinct.

Diagnoses comparisons

Comparing neurodegenerative conditions provides context between their estimated prevalences and their hospitalisation incidences. Hospitalisation rates were shown to be vastly different, reflecting inherent differences between the conditions. Rising hospitalisations for some conditions may be attributed to increased disease prevalence due to an ageing Australian population,¹⁴ diagnosis at earlier ages and improved long term outcomes,¹⁸ and improved longevity extending treatment periods and increasing periodic hospitalisations.⁴⁶ Further, all of the conditions being compared with post-polio are neurodegenerative progressive conditions. Some have faster rates of degeneration than others. Public AIHW data provides further clarity as to the hospitalisations by age group, with distinct trends among these specific neurodegenerative conditions.

Within this group, PD (annual mean 28 773) and MS hospitalisations (annual mean 31 212) have large incidences. For PD this is attributable to having the largest condition population yet a modest hospitalisation incidence (40 per 100), and for MS this is due to having a 1:1 population size to hospitalisation incidence ratio. MND had an annual mean of 3328, this constrained by the fast progression and early death of those with MND.¹⁵ Hospitalisations might increase if treatment options emerge extending the lifespan of those with MND. MG hospitalisations’ incidence increased three-fold in the 10-year period, capturing advancements in MG diagnosis and treatment approaches.⁴⁷

Those with LEoP/PPS had the lowest hospitalisation incidence of the conditions, at 2 per 100. Although comparable in estimated population to those with MS, those experiencing LEoP/PPS are difficult to identify and likely not fully captured in the obtained data. Factors include misdiagnosis, lack of diagnosis, low societal and clinician awareness of LEoP,¹ and psychosocial aspects (acute polio post-traumatic stress disorder, personality, dissociation, coping unpreparedness, false beliefs, health anxiety).^3,8

The majority of people living in Australia who have or are likely to develop LEoP are aged over 60.⁴⁸ Australians aged over 65 are more likely to be hospitalized.⁴⁹ The hospitalisation rate seen in this data appears incongruent with the estimated post-polio population size, health profile, and risk for hospitalisation, reinforcing concerns regarding population under-identification. On the contrary, Polio Australia’s existing prevalence estimate may be high. Capturing additional data points would improve the accuracy of post-polio prevalence, ideally through primary care, AIHW and ABS surveys, community screening, awareness campaigns, and a national patient registry. Further examination of Australia’s post-polio prevalence may inform worldwide estimates.¹²

Limitations of this study

The two AIHW data sources, while not compatible for combined analysis, together offered greater descriptive detail. Data likely contained individuals who were hospitalized more than once. ICD-10-AM coding changes, variability in facility-level code interpretation, and diagnosis errors may have created error in the data. The absence of hospitalisation data qualities denies commentary on such.

Conclusion

This study has provided a descriptive analysis identifying trends in the hospitalisation of people with post-polio conditions, and provided a data point informing post-polio disease prevalence in Australia. We identified the hospitalisation incidence of those with an existing post-polio diagnosis of B91 (late effects of polio) and G14 (post-polio syndrome) over a 10-year period. Significant trends were identified in annual, monthly, state, and capital versus regional analyses. This dataset provided new demographic detail on this post-polio population and contextual comparisons to four neurodegenerative conditions were established.

Acknowledgements

The authors would like to acknowledge the contributions of Dr Nigel Quadros who sadly passed away during the course of this research project. Dr Quadros was a well-respected member of the post-polio medical community in South Australia, Australia and world-wide, and ran the only post-polio clinic in South Australia (through The Queen Elizabeth Hospital). Dr Quadros led a number of research projects focused on post-polio as well as on sarcopenia in the general healthy and clinical populations.

Conflict of interest. None declared.

Data availability

The data underlying this article were provided by the Australian Institute of Health and Welfare (AIHW) under licence / by permission. Data will be shared on request to the corresponding author with permission of the AIHW.

Timothy J.H. Lathlean, Post-doctoral Research Fellow

Nigel Quadros, Clinical Consultant

Akhilesh K. Ramachandran, Doctoral Candidate

Michael J. Jackson, Clinical Educator