The average cost of pressure ulcer management in a community dwelling spinal cord injury population

Abstract

Pressure ulcers (PUs) are a common secondary complication experienced by community dwelling individuals with spinal cord injury (SCI). There is a paucity of literature on the health economic impact of PU in SCI population from a societal perspective. The objective of this study was to determine the resource use and costs in 2010 Canadian dollars of a community dwelling SCI individual experiencing a PU from a societal perspective. A non‐comparative cost analysis was conducted on a cohort of community dwelling SCI individuals from Ontario, Canada. Medical resource use was recorded over the study period. Unit costs associated with these resources were collected from publicly available sources and published literature. Average monthly cost was calculated based on 7‐month follow‐up. Costs were stratified by age, PU history, severity level, location of SCI, duration of current PU and PU surface area. Sensitivity analyses were also carried out. Among the 12 study participants, total average monthly cost per community dwelling SCI individual with a PU was $4745. Hospital admission costs represented the greatest percentage of the total cost (62%). Sensitivity analysis showed that the total average monthly costs were most sensitive to variations in hospitalisation costs.

Introduction

The Canadian Paraplegic Association estimates that annually approximately 35 individuals per million Canadians will experience a spinal cord injury (SCI) (1). Pressure ulcers (PUs) are a common secondary health complication amongst those with SCI 2, 3 with a prevalence of 27·8% in the Canadian community setting (2). This high rate of ulcers among the SCI population occurs most often over the ischial tuberosity as a result of extended wheelchair use (4). PUs develop where there is constant prolonged pressure over bony prominences, skin, muscle or soft tissue leading to damage of the affected areas (5).

Along with infection (5), death and decreased quality of life (6), PU is responsible for substantial health care costs to the system. A recent US study calculated the cost of a stage IV PU in US dollars to be close to $130 000 per hospital admission for a hospital‐acquired ulcer and $31 000 for each admission for a community‐acquired PU (7). These results were limited to the initial hospitalisation costs of a newly diagnosed PU. Less understood are the continuing costs after initial hospitalisation in community dwelling individuals with a PU, especially in a SCI population that has already incurred large health care expenses and are at high risk for PU. Allen and Houghton (8) determined that the cost for a SCI individual in a Canadian community setting with a stage III PU was $27 500 over 3 months (about $9000 per month of community care with electrical stimulation and an interdisciplinary wound team, including costs associated with health care professionals, rental equipment, supply use and loss in productivity). However, this cost was based on a case study of one patient; therefore, generalizability and representativeness of the results is limited.

The objective of this study is to determine the resource use and costs of a community dwelling SCI individual experiencing a PU from the Ontario societal perspective. This study will also examine costs by age, PU history, severity level, location of SCI, duration of current PU and PU surface area.

Methods

A cost analysis was conducted on community dwelling SCI individuals experiencing a PU and located in Toronto and London, Ontario, Canada.

Population cohort

The population cohort of SCI individuals was obtained from a pilot randomised controlled trial (RCT) comparing an interdisciplinary pressure management and mobility program (IPM) (consisting of occupational therapists and physiotherapist trained in wound care) to bed rest for the treatment of PU (ClinicalTrials.gov Identifier: NCT00796042). The active intervention of IPM is consistent with Canadian national (9) and provincial (10) best practice recommendations. However, many SCI individuals who develop a PU are managed with bed rest. At the time of enrollment, individuals were managing an existing PU with bed rest either as a personal decision or at the advice of their family physician or nurse. In this study, individuals were randomised to receive IPM or bed rest for 3 months, followed by a 4‐month period where they had the option to continue with current treatment or switch to another treatment option. Table 1 summarises the inclusion and exclusion criteria for the RCT.

Table 1.

Inclusion and exclusion criteria

Inclusion criteria	Exclusion criteria
• Adults (18–80 years) with a SCI resulting in quadriplegia or paraplegia	• Unable to achieve adequate pressure relief using available resources provided by research study
• Stage II–IV PU located anywhere on the body that has been present for at least 3 months and likely to heal over the 6‐month study period	• Stage IV deep pressure sore ± osteomyelitis that requires surgical closure
• Requires a wheelchair for locomotion	• Medically unstable or has any other medical condition that would prevent the individual from sitting in wheelchair for at least 2 hours (e.g. orthostatic hypotension)
• Is limiting their mobility because of concerns about skin condition (bed rest)	• Has a limited life expectancy that is not sufficient to complete the study
• Able to undergo detailed wound and medical assessment that includes blood analysis and wound culture
• Provides consent to access relevant medical records
• Able to adhere with study protocol, pressure management and mobility program, wound care protocol, and evaluation procedures for 8‐month study period

SCI, spinal cord injury; PU, pressure ulcers.

Perspective

Costs were examined from the societal perspective (including costs to the health care system, social assistance and labour sector), incorporating the economic impact to the Ontario Ministry of Health and Long‐term Care, patients and informal caregivers.

Resource utilisation

Medical care resource utilisation was collected from participants during the study period through electronic questionnaires. Data collected included: (i) emergency room (ER) visits, (ii) inpatient hospitalisation stays, (iii) general practitioner visits, (iv) physician specialist visits,(v) clinic visits, (vi) nursing and allied health practitioner visits (e.g. physiotherapist, occupational therapist, community nurse time, wound specialist visits), (vii) laboratory tests, (viii) diagnostic imaging (e.g. x‐rays), (ix) drug use (e.g., prescription and over‐the‐counter), (x) paid employment time lost, (xi) homemaking, or volunteer time lost and (xii) assistance with daily activities time (unpaid). For family physician visits, physician specialist visits, nursing and allied health providers, diagnostic and laboratory tests and assistance with daily activities, participants reported whether the resource use was specific to PU. Thus, PU attributed resource use could be analysed.

Costs

The unit costs were collected from publicly available sources (e.g. Ontario Case Costing Initiative, Ontario Physician Schedule of Benefits, Ontario Drug Benefit Program) and published literature. Physician costs, laboratory tests and medical procedure unit costs were obtained from Ontario Ministry of Health and Long‐term Care Schedule of Benefits. Hospitalisation costs for surgical procedure were obtained from the Ontario Case Costing initiative. Hospitalisation costs for emergency room visits or medical care clinic were based on the results by Zoutman and colleagues (11). Average hourly wages for allied health care professionals, assistance for the individual and homemaking, and primary caregiving time off was obtained from Human Resources and Skills Development Canada. For nurses and allied health professionals, the wages were increased by 13% to account for benefits (12). The average Canadian wage reported by Statistics Canada was used to place a monetary figure on lost volunteer time. This will be further explored in the sensitivity analysis. Unit costs for drugs were obtained from the Ontario Drug Benefit Formulary, or Saskatchewan drug plan formulary when not available in Ontario. Table 2 summarises the unit costs used in this analysis. Costs are presented in 2011 Canadian dollars. Where Ontario costs were not available, provincial or national estimates were used. Cost estimates reported in earlier years were inflated to 2011 using the Consumer Price Index health and personal care rate published by Statistics Canada (13). Total costs for each item was calculated by multiplying the total number accrued for each item, by the corresponding unit costs of each item. Since homemaking, volunteer and primary caregiver time lost was reported in days, but unit costs reported per hour, it was assumed that an individual lost 2 hours of homemaking time per day, 4 hours of volunteer time per day and 8 hours primary caregiving time per day. Time spent in the emergency ward of a hospital was not collected, thus the total average time was assumed to be the total average emergency department wait time as reported on the Government of Ontario website (14).

Table 2.

Unit costs

Variable	Unit cost	Definition	References
Health care professionals
Family physician	$35·40	Code A004	(15)
Gastroentologist	$148·95	Code A415	(15)
Surgeon	$90·30	Code A035	(15)
Respirologist	$148·95	Code A475	(15)
Cardiologist	$148·95	Code A605	(15)
Ear, throat, nose specialist	$76·00	Code A245	(15)
Allergist	$70·25	Code A025	(15)
Infectious disease specialist	$148·95	Code A465	(15)
Wound specialist	$70·25	Code A025	(15)
Physiatrist	$170·70	Code A315	(15)
Radiologist	$35·70	Code A335	(15)
Physiotherapist	$36·38	NOC code 3142 Ontario 2005. Wage inflated to 2011 dollars including 13% benefits	(16)
Occupational therapist	$36·48	NOC code 3143 Ontario 2005. Wage inflated to 2011 dollars including 13% benefits	(16)
Community nurse	$40·87	NOC code 3152, 2009. Wage inflated to 2011 dollars including 13% benefits	(16)
Wound care specialist	$40·87	NOC code 3152, 2009. Wage inflated to 2011 dollars including 13% benefits	(16)
Personal support worker	$17·74	NOC code 6471 visiting homemakers, housekeepers and related occupation, 2009. Wage inflated to 2011 dollars including 13% benefits	(16)
Dietician	$34·82	NOC code 3132 British Columbia 2007, Ontario not available. Wage inflated to 2011 dollars including 13% benefits	(16)
Laboratory tests and diagnostic imaging
Complete blood count	$8·27	Code L393	(17)
Helicobacter pylori	$12·93	Code L628	(17)
24 pH	$3·62	Code L034	(17)
Gastric emptying	$155·65	Code J829	(15)
Upper gastrointestinal	$113·00	Code X109	(15)
Upper endoscopy	$161·31	Code Z399	(15)
CT chest	$68·35	Code X406	(15)
CT abdomen	$91·15	Code X409	(15)
Esophageal motility	$89·45	Code G350	(15)
Electrocardiography	$16·65	Code G310 and G313 Electrocardiogram	(15)
Pulmonary function	$17·40	Code J301 $7·85 professional	(15)
Chest x‐ray	$33·75	Code X091	(15)
Back x‐ray	$34·90	Code X028	(15)
Urine culture and sensitivity	$3·62	Code L641	(17)
Change kidney stents	$105·25	Code Z623	(15)
Teeth x‐ray	$33·15	Code X006	(15)
Chest ultrasound	$82·30	Code J125	(15)
Wound swab	$12·93	Code L628	(17)
Myelogram	$24·45	Code X006	(15)
Abdomen x‐ray	$33·10	Code X101	(15)
Loss in productivity and out of pocket expenses
Homemaker days off	$15·72	NOC code 6471. Wage inflated to 2011 dollars	(16)
Volunteer days off	$22·81	Average wage October 2010	(16)
Primary caregiver days off	$12·37	NOC code 6474. Wage inflated to 2011 dollars	(16)
Average Canadian wage	$22·81	Average wage October 2010	(16)
Assistance in health care related activities	$36·17	NOC code 3152. Wage inflated to 2011 dollars	(16)
Assistance in personal care related activities	$15·70	NOC code 6471. Wage inflated to 2011 dollars	(16)
Assistance in house care related activities	$15·70	NOC code 6471. Wage inflated to 2011 dollars	(16)
Assistance in transportation related activities	$10·66	NOC code 7413. Wage inflated to 2011 dollars	(16)
Hospitalisation
Loop colostomy – hospitalisation cost (less physician fees)	$50 337·38	36 days at $1334 per day. Per diem cost calculated from OCCI report of $25 349 for 19 days mean length of stay – acute inpatient. Total cost inflated to 2011 dollars	(18)
Colostomy – physician procedure cost	$575·81	Code S157	(15)
Most responsible physician follow‐up – colostomy recovery	$1201·30	Code C413, C122, C123, C124, C412, assuming one follow‐up per day	(15)
Debridement of PU – hospitalisation cost (less physician fees)	$29 934·77	32 days at $892·47 per day. Per diem cost calculated from OCCI report of $15 172·00 for 17 days mean length of stay – acute inpatient. Total cost inflated to 2011 dollars	(18)
Debridement Physician procedure cost	$228·78	Code Z144	(15)
Most responsible physician follow‐up – debridement recovery	$1052·80	Code C023, C122, C123,C124, C412, assuming one follow‐up per day	(15)
Failure to thrive – hospitalisation cost (less physician fees)	$5784·65	7 days at $788·40 per day Per diem cost calculated from OCCI report of $23 652 for 30 days mean length of stay is $788·40/day – acute inpatient. Total cost inflated to 2011 dollars	(18)
Hospitalisation
Most responsible physician follow‐up – failure to thrive recovery	$337·40	Code C133, C122, C123, C124, C412, assuming one follow‐up per day	(15)
Emergency visits
ER visit per hour	$326	Cost was inflated to 2010 dollars	(11)
Drugs
Apo ferrous gluconate	$0·02	300 mg Tab	(19)
Tetracycline	$0·05	250 mg Cap	(19)
Ciprofloxacin	$1·25	500 mg Tab	(19)
Ramipril	$0·38	5 mg Tab	(19)
Oxycocet	$0·01	5 mg and 325 mg Tab	(19)
Acetaminophen	$0·01	325 mg Tab	(19)
Amoxicillin	$0·34	500 mg Tab	(19)
Avelox	$5·58	400 mg Tab	(19)
Azithromycin	$2·47	250 mg Tab	(19)
Cephalexin	$0·45	500 mg Cap	(19)
Clavulin	$0·44	250 mg and 125 mg Tab	(19)
Doxycycline	$0·51	100 mg Cap	(20)
Saskatewan cost, Ontario not available
Ferrous fumarate	$0·12	300 mg Cap	(19)
Levofloxacin	$5·34	500 mg Tab	(19)
Macrobid	$0·67	100 mg Cap	(19)
Mycostatin powder	$0·06	100 000 U/ml O/L	(19)
Novamoxin	$0·34	500 mg Cap	(19)
Oxycodone	$0·85	10 mg Tab	(19)
Senokot	$0·06	8·6 mg Tab	(19)
Soflax	$0·04	100 mg Cap	(19)
Trimethoprim	$0·19	100 mg Tab	(19)
Zythroma	$2·47	250 mg Tab	(19)
Zyvoxam	$70·64	600 mg Tab	(19)

CFap, capsule; CT, computed tomography; ER, emergency room; IV, intravenous; NOC, national occupational classification; Tab, tablet; PU, pressure ulcers.

Analysis

After completion of the RCT, it was clear that there were no significant differences between the amount of activity, equipment allocation, number of hours spent in bed, and wound healing outcomes between the two groups. Furthermore, many logistical issues were encountered that interfered with implementation of the community‐based IPM program. Therefore, resource utilisation from both study groups have been combined for the purpose of this cost analysis.

The total average monthly cost per community dwelling SCI individual with a PU was determined by multiplying the resource use reported by the unit cost of each resource, summing the total cost per month then calculating the mean cost per month. Average costs were also stratified by age (above/below 65 years), PU history (any previous PU), severity level (based on stage at baseline), location of SCI (cervical or thoracic), duration of current PU (greater/less than one year) and PU surface area (greater/less than 10 cm²). Stratification points for SCI site, duration of current PU, PU surface area and location of SCI were determined based on the clinical opinion of one of the authors (PH) a priori. Average costs were determined for individuals above and below age 65 years based on Chen et al.(21) who found PU are more common among SCI individuals above 65 years of age. PU severity level was assigned after direct examination of the wound using the National Pressure Ulcer Advisory Panel staging system, 2007 (22).

Descriptive statistics such as mean and standard deviation were used to summarise the results. Microsoft^® Excel 2007 was used to compute the results and not subjected to statistical analyses given the small sample.

Sensitivity analysis

A sensitivity analysis was conducted on the variables that were considered to have the most uncertainty because of the assumptions required to calculate a cost: (i) number of hours assumed to be lost due to a day off for homemaking, volunteering and care‐giving was ranged between 0 and 8 hours a day; (ii) cost of inpatient hospitalisation was varied between the minimum and maximum costs obtained by the Ontario Case Costing Initiative; (iii) the entire average wait time in the emergency department, as reported in Ontario, was assumed to be representative of the entire time the individual was occupying a patient bed in the emergency department and (iv) finally, the cost of assistance provided from friends and family (unpaid) and the time lost for home making, volunteering and care‐giving was also excluded in a sensitivity analysis to explore the impact of monetising unpaid time in our analysis

Ethical considerations

Informed consent was sought and obtained from all participants in this study. The study was approved through the University of Western Ontario Human Research Ethics Committee.

Results

In total, 14 individuals consented and participated in the pilot study. Two individuals did not complete any of the medical care resource forms throughout the entire study period and thus were excluded from this analysis.

The average age amongst the 12 individuals was 52·4 years (±15·2) with a median (range) of 53(24–70), 67% were male. The cohort had an average of 21 years with SCI with a median(range) of 22(1–48) (42% quadriplegia, 50% paraplegia, 8% unknown site of injury) and 67% had experienced a previous of PU. The average duration of time with the current PU was 25 months with a median(range) of 8·5(2–144), where 8% had a stage II PU, 67% had a stage III PU and 25% had a stage IV PU. The average PU surface area of wounds was 22 cm² with a median(range) of 10(0·1–70·5) and the average depth was 3 cm with a median(range) of 3·1(0·4–4·5). The study demographics are presented in Table 3.

Table 3.

The study demographics

	Community dwelling spinal cord injured individuals (n = 13)
Mean age ± standard deviation (SD) (median, range)	52 ± 15 (53, 24–70)
% male	67%
Mean years with spinal cord injury ± SD (median, range)	21 ± 16 (22, 1–48)
Cervical site injury	33%
Thoracic site injury	58%
Unknown site of injury	8%
% with a previous pressure ulcer	67%
Mean duration of current pressure ulcer (months) ± SD (median, range)	25 ± 41 (8·5, 2–144)
% stage II pressure ulcer	8%
% stage III pressure ulcer	67%
% stage IV pressure ulcer	25%
Mean wound surface area (cm2) ± SD (median, range)	22 ± 25 (10·2, 0·1–70)
Mean wound size depth (cm) ± SD (median, range)	3 ± 1 (3·1, 0·4–4·5)

Total average monthly cost per community dwelling SCI individual with a PU was $4748 (±$9279 standard deviation) in 2011 Canadian dollars.

Average monthly costs were stratified into different categories (Figure 1). Hospitalisation costs contribute to the greatest percentage of the cost. In total, two patients (17%) were admitted three times as a hospital inpatient with an average length of stay of 25 days (± 15·7, range 7–36 days). Five patients (42%) had seven visits to the hospital emergency department. After hospitalisation, cost for nursing and allied health care providers (physiotherapy, occupational therapy, wound care nurse), and costs for assistance contributed to the greatest proportion of costs. The majority of costs within nursing and allied health care providers originated from the community nurse, accounting for 59% of costs in this category (Figure 2). Costs were stratified per month through the observation period (Table 4).

Figure 1.

Percentage of average monthly costs separated in cost categories.

Figure 2.

Percentage of nursing and allied health care provider costs stratified by profession.

Table 4.

Number of individuals reporting medical care resources per month and monthly cost per patient separated by costs to the Ontario Ministry of Health and Long‐term care and the individual/caregiver

	Month 1	Month 2	Month 3	Month 4	Month 5	Month 6	Month 7
Number followed up	12	10	10	7	6	4	6
Cost per patient (standard deviation)	$8595 ($16424)	$2510 ($2446)	$5905 ($11126)	$3959 ($4315)	$3086 ($2322)	$3383 ($2381)	$2342 ($3270)

Average cost for each month throughout the observation period and total average monthly cost for individuals stratified by stage of PU severity, location of SCI injury, age, duration of current PU, surface area PU and whether the individual has had a previous PU are presented in Table 5.

Table 5.

Average total monthly cost for individuals stratified by different criteria

	Average monthly cost ± standard deviation (min, max)
Stage of pressure ulcer
Stage II pressure ulcer (n = 1)	$683 ±$636 ($10, $1273)
Stage III pressure ulcer (n = 8)	$6098 ±$10403 ($35, $58,170)
Stage IV pressure ulcer (n = 3)	$823 ±$1584 ($35, $5,346)
Level of injury
Cervical level injury (n = 5)	$5303 ±$4203 ($214, $19214)
Thoracic level injury (n = 6)	$1217 ±$1888 ($10, $8719)
Age
Less than 65 years (n = 8)	$3621·53 ±$4023 ($10, $19 214)
Greater than 65 years (n = 4)	$7265 ±$15 601 ($35, $58 170)
Duration of pressure ulcer
Duration of pressure ulcer <1 year (n = 8)	$5734 ±$10 499 ($35, $58 170)
Duration of pressure ulcer >1 year (n = 4)	$1860 ±$2531 ($10, $8719)
Size of pressure ulcer
Surface area of <10 cm2 (n = 6)	$3641 ±$4243 ($35, $19 214)
Surface area of >10 cm2 (n = 6)	$6539 ±$14 046 ($10, $58 170)
History of pressure ulcer
Previous pressure ulcer (n = 8)	$1667 ±$2038 ($10, $8719)
No previous pressure ulcer (n = 4)	$9736 ±$13 540 ($1846, $58 170)

When stratified by PU severity level, stage III had greater average monthly costs compared to stages II and IV. When categorised by level of SCI, cervical level injury had five fold greater average monthly costs compared to those with thoracic level injury. With respect to age, those who were 65 years and older had double the monthly costs of those who were under 65 years. If the duration of the PU was <1 year, the costs were almost threefold higher in comparison to those who had the PU for more than 1 year. Having a larger size of the PU (greater or < 10 cm²) almost doubled the costs. Lastly, having no history PU led to approximately six times the average monthly costs compared to those who had previous history with PU.

The total and percentage of average monthly costs associated with PU as reported by the study participants are presented in Table 6. Sixty‐three percent of the total costs were attributable to PU; with the majority of the costs being associated with nursing and allied health care provider visits.

Table 6.

Total and percentage of costs associated with PU

	Family physician cost	Physician specialist costs	Nursing and allied health care provider costs	Diagnostic and laboratory test costs	Assistance with daily activities costs	Total costs
Pressure ulcer related	$3	$41	$952	$2	$234	$1232
Total average cost	$23	$52	$1120	$10	$747	$1952
Percentage of costs associated with pressure ulcer	21%	78%	85%	17%	31%	63%

PU, pressure ulcers.

For the sensitivity analysis, it appears that there was little change in the average monthly cost per individual when varying the unpaid time lost for homemaking, volunteering and primary caregiver (Table 7). This was also observed when assuming complete ER wait time was representative of the time an individual occupies a hospital bed in the ER. However, the changes in average monthly costs were more substantial with changes to hospitalisation costs. There was also a greater than $1000 decrease in average monthly costs when unpaid costs were excluded from the calculation.

Table 7.

Changes to average monthly cost per individual resulting from changes in parameters with uncertainty

Variable changed	Average monthly cost per individual ± standard deviation (min,max)
No costs for unpaid time lost for homemaking, volunteering and primary caregiver	$4331 ±$8987 ($10, $57 226)
Unpaid time lost reduced to 2 hours for homemaking, volunteering and primary caregiver	$4626 ±$9233 ($10, $58 170)
Unpaid time lost increased to 8 hours for homemaking, volunteering and primary caregiver	$5458 ±$10 107 ($10, $61 000)
Minimum hospitalisation costs for inpatient visits as reported by Ontario Case Costing Initiative	$3416 ±$3845 ($10, $19 214)
Maximum hospitalisation costs for inpatient visits as reported by Ontario Case Costing Initiative	$18 025 ±$84 293 ($10, $586 071)
Complete emergency room wait time representative of time individual occupies hospital bed in emergency room	$4936 ±$9724 ($10, $60 597)
No costs for unpaid assistance from friends and family and unpaid time lost	$3587 ±$8799 ($0, $56 934)

Discussion

To our knowledge, this is the first study to evaluate the costs associated with a community dwelling SCI cohort with PU in Ontario, Canada from a societal perspective. In our study, we observed a cost of over $4700 per individual per month. Nursing‐related care had a significant impact, representing 59% of total nursing and allied health care personnel costs. Considering the subjects had mobility restrictions compounded by a PU that is likely due to wheelchair use, access to rehabilitation professionals (Occupational and Physical Therapy) was extremely low, representing 2% and 0·5% of total nursing and allied health care personnel costs, respectively. Overall, the total cost attributable specifically to PU as reported by the study participants was 63% of the total cost reported by the cohort.

In comparison to other studies such as Allen and Houghton (8) who estimated that the cost for a stage III PU in a community dwelling SCI case was $27 500 for 3 months, or $9000 per month Canadian dollars (2004), the average monthly costs were lower in our study. Though the average costs were observed to be higher despite the lack of hospitalisation during the study duration, the Allen et al.(8) study included costs that were not collected in our study (lost wages ($4660), wound care supplies ($700) and equipment rental costs ($1550).

At the time of recruitment, individuals in this study had their ulcers present for an average of 25 (±41) months before the commencement of data collection. Using the monthly cost estimated in our study, the average cost of PU management at the start of the study was $117 500 per individual ($4700 × 25 months). To manage the entire study cohort of 12 individuals, the total cost would be close to $1·4 million ($117 500 × 12 individuals). Only one individual in the study sample experienced complete ulcer healing during the study period. Thus, the costs continue to accumulate post study period.

According to US figures, approximately 87·7% of surviving individuals hospitalised for SCI are discharged back to the community (23). Assuming that rates are similar in Canada and consistent over time, and with an estimated 36 000 SCI individuals living in Canada (24), there may be approximately 31 500 SCI individuals living in the community. Therefore, with a prevalence of PU ranging from 11·5 to 21% depending on the individual's age or year since injury (21), the total annual cost of community dwelling SCI individuals with PU in Canada from the societal perspective would be estimated to be between $173 and $316 million.

Our study confirms previous reports of high medical resource usage among SCI individuals in Alberta compared to a non‐SCI population (25). Also, in a cross‐sectional analysis of Model Spinal Cord Injury Systems centers in the US, Cardenas et al.(26) found PU to be one of the most common reasons for re‐hospitalisation in SCI individuals as long as 20 years post incident. Furthermore, PUs in SCI individuals contribute to a disproportionately higher length of stay in the hospital 27, 28.

In the sensitivity analysis, it was observed that the results of this study varied drastically when hospitalisation costs were varied. The base‐case analysis yields a total average monthly cost per community dwelling SCI individual with a PU of $4748. Using the minimum and maximum hospitalisation costs, the range was between $3400 and $18 000. However, it was noted that the maximum hospitalisation costs were likely based on a maximum hospital length of stay that is much larger than that reported in our study cohort.

Individuals with stage III PUs had higher monthly costs compared to the more severe stage IV PUs. This observation may be a result of the very small number of individuals with stage IV PU (n = 3) not being representative of individuals with this level of severity in general. Individuals with cervical level injury also consumed greater average costs compared to thoracic level injury. This may not be surprising given that individuals with cervical injury are tetraplegic and will require more assistance and health care than a paraplegic. Greater resource use is also expected for older individuals and this was observed when average costs were stratified for individuals younger and older than 65 years. Individuals reporting that their current ulcer started less than a year ago also reported higher average costs compared to current ulcers greater than a year. As one would expect, average monthly costs were higher for individuals with a PU with a greater surface area. Finally, individuals with no prior history of PU had higher costs compared to individuals with past PU. It is likely that individuals experiencing their first PU are likely to demand greater medical attention and consuming more resources compared to individuals with repeated PUs. The sample size of this cohort is small and the variability of monthly costs is large. Thus, results were not analysed statistically and differences observed in the stratified sub‐groups should be interpreted with caution. This overall analysis was based on a small sample size with large variability. A larger number of patients are needed to reduce the variance and confirm the results of this study. Therefore, the results should be interpreted with caution, especially comparison of costs between different categories of stratification, where sample sizes are even smaller. However, in practicality, it is difficult to recruit a large number of community dwelling SCI individuals willing to be followed up prospectively for a lengthy period of time. Thus, the data presented here is an important contribution to SCI and PU research.

There are several limitations to our study. First, individuals entered the pilot study several months after experiencing a PU. Therefore, any unique initial PU treatment costs were not captured. Alternatively, none of the participants had their PU completely healed by the end of the follow‐up period. Therefore, any reduction in costs near the end of the PU was not recorded. It is assumed that the period of follow‐up in this pilot study represents the time between the initial treatment of PU and complete healing of PU, thus excluding the initial and end costs associated with PU. Second, the study relied on self‐reported questionnaires, which resulted in missed follow‐up by some participants at different time points within the duration of the data collection period. Also, costs paid for by the individual were not systematically disaggregated from the patients in the questionnaire. Therefore, out‐of‐pocket cost was not reported in this analysis. Moreover, self‐reported questionnaires are subject to recall bias by the participants. The data should be interpreted cautiously with this limitation in mind. Third, unpaid leisure, homework and education time lost was not included in the questionnaire and thus could not be incorporated into the average cost. Thus, the average cost presented in this study is likely an underestimation. Fourth, the cost of nursing and allied health practitioner benefits may be an underestimate. With the inclusion of vacation and sick days, the calculated wage including benefits would be increased by an additional 13% (29). Fifth, wound care training costs, wound care equipment, wound care supplies and equipment associated with diagnostic imaging were not collected and thus not included in the overall costs. Therefore, the costs presented in this analysis may be an underestimation of the true costs for SCI individuals with PU. A prior study exploring resource use for wound care in a Canadian health care setting has observed wound care equipment and supply cost of $351 (±$473 standard deviation) (8). Incorporating this value to the estimation in our study would result in an average monthly cost of $5096. Sixth, the results of this study are based on the health care resources that were collected in the patient questionnaire. Although most of the costly resources were included, there may have been items that have been missed. The result of this would be an underestimation of the cost for SCI individuals with PU. Finally, Table 5 presents lower stage IV average costs in comparison to stage III average costs. This may be due the small sample size of only three individuals who had stage IV PU in comparison to the eight individuals who had stage III PU. Therefore, the stage IV average costs presented in this study may not be representative of the true costs for individuals with this level of PU severity.

Despite the limitations of the study, the results of this analysis present an estimate of total costs of community dwelling SCI individuals with PU. This study also highlights the significant impact of nurses to the care of PU. Woodbury et al.(30) state that PUs are a significant health care concern that is under‐researched and that there is a lack of awareness about resources dedicated to educating healthcare professional on the prevention, treatment and management strategies associated with PU. Our study adds to this underdeveloped research area and overall, quantifies the societal cost of a community dwelling SCI individual with PU. The results of this analysis present the substantial costs incurred by this population and suggest a need for better preventative measures and treatment of PU in a group that already requires significant health care resources. This study provides useful information for decision makers assessing the health economic impact of PU in SCI individuals. Future costing studies of PU in an SCI population should focus on addressing some of the limitations of this study. For instance, a comparative study with an SCI cohort without PU may provide a more accurate estimate of PU attributable costs. Collection of health care resource utilisation by using administrative data may be a means to extracting data over a longer period of time without missing data points.

Acknowledgement

This work is attributable to School of Physical Therapy, University of Western Ontario, London, Ontario, Canada.