School of Population Health Tamaki Campus University of Auckland Private Bag 92019 AUCKLAND
Cost effectiveness of impact absorbent flooring in reducing fractures among institutionalized elderly
February 2008
Frida Njogu Paul Brown * School of Population Health University of Auckland * For more information, contact: Paul Brown School of Population Health University of Auckland Private Bag 92019 Auckland, NZ
[email protected]
1 of 17
Executive Summary •
The rate of falls and subsequent injuries among the elderly in institutional care is on the rise, and results in significant health resource use. Some injuries can be prevented by the use of external hip protectors or by impact absorbent flooring.
•
The purpose of this report is to assess the cost-effectiveness and cost-utility of two strategies for preventing injuries in residential facilities – provision of hip protectors to elderly residents or installing impact absorbent flooring in the facilities.
•
A decision tree model was constructed from New Zealand and international studies showing the probabilities of preventing injury from two strategies: o Providing hip protectors to all residents of residential homes; and o Installing impact absorbent flooring
•
Costs and cost effectiveness of each strategy was calculated. Outcome measures were hip fractures averted and any injury averted for the cost effectiveness analysis; and QALYs gained from averted hip fractures for cost utility analysis.
•
The results suggest that for a cohort of 10,000 elderly people, providing hip protectors averts 26 hip fractures at an additional cost of $1,057,384. This equates to a cost per hip fracture avoided of $44,058 and a cost per QALY of over $220,000.
•
Installing absorbent flooring averts 63 hip fractures and an estimated 800 other types of fractures. It is estimated to reduce overall costs by $411,093, meaning that it leads to better outcomes and saves money.
•
Sensitivity analysis suggests that the results are sensitive to the assumed efficacy of the new flooring in preventing fractures. For hip protectors, the results are sensitive to the assumed rate at which older people are wearing hip protectors at the time of a fall.
•
Impact absorbent flooring may be both less costly and more effective than providing hip protectors. Further study is needed to confirm the rate of injury when impact absorbent flooring is used in residential facilities.
2 of 17
1. Introduction Falls are a major cause of morbidity and mortality among the elderly, accounting for 84% of blunt trauma in this age group [1]. With the projected increase in the elderly population in New Zealand, the burden of injuries due to falls on health services is expected to rise.
The human body has natural mechanisms to prevent falls, or in the event of a fall, to mitigate the severity of the resultant injury. These natural defenses include good eyesight, postural balance, sufficient adipose tissue for cushioning, elasticity and deformability of tissues and reflexes among others, most of which are deficient among the elderly [2, 3]. The elderly, especially women, are also at the highest risk of osteoporosis, hence the higher incidence of fractures upon falling. This is worse among those residing in residential facilities as they tend to have poorer health than community dwellers. About half of the elderly living in residential homes fall at least once a year [4-7], and between a third and two thirds of these sustain injuries [4, 5, 7-9]. The injuries from falls include hip fractures, vertebral fractures and head injuries, with hip fractures being the most severe and costly of these injuries.
The occurrence of injuries can be analyzed using the epidemiological triad of the host, vector and environment. In the case of falls resulting in injuries, the host is the elderly person, the environment the surface on which they fall, and the agent the fall and the associated energy [10]. Most strategies to reduce falls and/or subsequent injuries have focused mostly on host factors e.g. osteoporosis by use of HRT or other medications and some on the agent i.e. energy of the fall e.g. by use of external hip protectors. Few have looked at the environment of the fall, i.e. the floor, or the manner in which the floor acts as an agent by absorbing the impact. Those studies focusing on the surfaces coming into contact have mainly been on hip protectors, to help reduce the impact of the force upon contact. Hip protectors, however protect only from hip fractures, and only when the person falls sideways yet only 30% of falls occur sideways; another 25% occur backwards and 14% forwards [7]. Though effective in reducing hip fractures [6, 11-13], compliance remains a major problem, with rates of 37-76% [6, 12, 14, 15]. This has
3 of 17
necessitated the development of a strategy that is passive i.e. requires no compliance on the part of the elderly person. A type of flooring material that has dual impact absorption, i.e. is firm to walk but deforms when a lot of pressure is applied (e.g. on a fall) to reduce the peak impact hence less risk of injury would be ideal [16, 17]. This has been developed and found to be effective in reducing peak force on impact [18].
As most (95%) injurious falls among those who dwell in residential homes occur indoors [7], the use of impact absorbent flooring would have a significant effect on injury risks. In addition, the floors will help reduce other fractures that are not catered for by hip protectors. However, the cost of replacing the conventional floors with the impact absorbent flooring is high.
This paper analyzes the cost effectiveness and cost utility of the impact absorbent flooring in reducing hip fractures and other injuries. The study compares the costs, outcomes and cost effectiveness of two strategies for preventing injuries compared to the current (do-nothing) approach: i) Providing hip protectors to all residents and ii) using impact absorbent flooring.
For the first strategy, it is assumed that elderly residents are provided with 3 sets of hip protectors each year. Hip protectors have been shown to be effective in reducing hip fractures resulting from falls. However, hip protectors are only effective if worn, and one of the major issues with providing hip protectors is the rate of compliance by elderly residents. In addition, a hip protector strategy requires training staff in residential homes and on-going administration of the program to be effective.
Impact absorbent flooring has two advantages over hip protectors. First, there are no issues with compliance since the flooring provides uniform protection to everyone inside the facility. Second, it protects against injuries to the head and other areas, not just hip. Although hip fractures are associated with the greatest morbidity and mortality, fractures and injuries to other areas can still be substantial.
4 of 17
Impact absorbent flooring is more expensive than traditional flooring and thus the cost effectiveness of this option is uncertain. The analysis below examines the case where the absorbent flooring is being installed either in a new facility or when the flooring of an established facility is being replaced. Because absorbent flooring does not require any additional installation, the analysis considers only the additional cost of the flooring and not the cost of installation (since that will be equivalent in each strategy).
2. Methods Using a decision tree model, the cost effectiveness and cost utility of using the impact absorbent flooring to prevent injuries among a hypothetical cohort of 10,000 elderly persons in residential homes over a one year period was examined. Data was obtained where possible from New Zealand studies/literature, though the majority comes from international literature.
2.1 Size of facility Figures from the Ministry of Health (2002) suggest that there are 34,157 beds in residential care facilities. This includes 23,412 (69%) in Rest Homes, 9559 (28%) in Private Hospitals and 1186 (3%) in dementia units. The size of the facilities varies across New Zealand. For the purposes of this study, the analyses are based on representative facility of 50 beds.
2.2 Outcomes measures: Hip Fractures Avoided and QALYs Because hip fractures are associated with significant morbidity and mortality, reductions in hip fractures can be seen as representing an increase in quality of life for elderly people. Quality adjusted life years are one way to measure the reduction in life expectancy and quality of life associated with these hip fractures. Previous studies have shown that hip fractures are associated with a utility loss of 0.2 QALYs across different ages [23, 24]. Both measures will be used in the analysis.
2.3 Risk of falling and injury
5 of 17
The risk of falling among institutionalized elderly persons in various countries has been found to range from 30-70% [4-7, 19, 20], usually higher than among community dwellers, while the risk of injury upon a fall ranges from 30% to 63% [4, 5, 7-9], though different studies include different types of injuries. For the base case analysis, 50% risk of falling was used, and varied at 30-70% in the sensitivity analysis. The risk of sustaining a hip fracture after a fall is 1-2% [5-7], head injury 19.3% [8], any fracture (including hip) 6.1-9% [4, 5]. A 2% hip fracture rate was used in the base case analysis and varied at 1% in sensitivity analysis, while a 7.5% rate for all fractures was used and varied at 6 and 9%.
2.4 Compliance and rates of injury: Hip protectors The compliance rate with hip protectors varies greatly, and declines with time after introduction. At 6 months, compliance rate ranges from 45% to 48% [14, 15], while at one year it is 37% [15]. Other studies show an overall compliance rate of 64% , and a wearing rate at time of fall of 42-76% [6, 12]. In the base case analysis, a 48% compliance rate was used, as this was based on a New Zealand study in which a cohort of institutionalized elderly individuals were followed up over an 18 month period; the figure will be varied at 37 and 76% in the sensitivity analysis.
The odds ratio of sustaining a hip fracture if one falls while wearing a hip protector as compared to those not wearing one is 0.31 - 0.77 [6, 11-13]. An odds ratio of 0.5 is used in the base case analysis, and varied at 0.3 and 0.77. Carpeting was found to reduce the risk of injury by 80% and 63% upon falling in different studies [21, 22].
2.5 Risk of injury: Absorbent flooring The impact absorbent flooring has been found to reduce peak force on falling by 65-80% compared to bare wood or concrete, and by 20-40% compared to wood or concrete floors overlaid with a carpet and underlay [18]. The extent to which this reduces injury is currently being examined, but preliminary evidence suggests it may reduce injuries by 2/3. Thus, for the purposes of this analysis, it is assumed that the flooring will be 63%
6 of 17
effective in reducing injury resulting from falling. In the sensitivity analysis, rates of 30% and 80% reduction are used.
2.6 Cost parameters 2.6.1
Injury from falling
The cost of treating a hip fracture was $23,859 as reported by Brown et al [24] was obtained from New Zealand public hospitals. This is inclusive of inpatient and rehabilitation costs. The cost of treating a person with a head injury was derived from a Finnish study, in which the average cost of treating injuries was €944 ($1888), and half of all injuries were head injuries [8]. This figure was used in the sensitivity analysis, and varied between $500 and $5000 as costs of treating various head injuries vary greatly.
Treatment costs of other fractures were obtained from a New Zealand study [24], in which treatment costs of vertebral and other osteoporotic fractures were reported, and a formula used to determine average costs of treatment a as $1,687 for ‘other’ fractures (including vertebral).
2.6.2 Cost of absorbent flooring The cost of the new flooring is expected to be about twice the cost of conventional flooring, at $150 per sq. meter compared to $75 for the current [25]. Labor costs of replacing the tiles would not be included, as it was assumed that the new tiles will be laid only when the old tiles need to be replaced, or at the construction of new buildings hence equal labor costs. The cost of the current flooring will not be included, as it is common to both the current and hip protector strategies; and only the cost difference is considered when comparisons are made with strategy II.
a
Only 8% of vertebral fractures are hospitalized. Cost of hospitalization is $12,450 per fracture, hence average cost incurred is 0.08*12,450 = $996. Vertebral fractures comprise 33% of all fractures (Brown et al, 2007) Cost of treating ‘other’ fractures: these comprise 62% of all fractures, and include such fractures as rib, forearm fractures, etc. Average cost of treatment is $2,191. Overall average cost of treating other fractures (including vertebral) is (0.33*996)+(0.62*2,191) = $1,687.
7 of 17
Table 1: Epidemiological and cost parameters for base case and sensitivity analysis Variable
Figures from literature
Base case
Probability of fall among elderly
30% [19, 20], 40% [7], 45% [4], 49% [5], 2970% [6] 30% [8], 34% [9], 45% [5], 45.8% [4], 63% [7] 1% [7], 1.97% [6], 2% [5] 19.3% [8]
50%
Sensitivity Sensitivity (minimum) (maximum) 30% 70%
2%
1%
6.1% [4], 9% [5]
7.5%
6%
9%
At 6 months: 45% [15], 48% [14] At 1 year: 37% [15] Others: 64% [6] Wearing rate at time of fall: 42% [6], 76% [12] 0.31 [11], 0.4 [12], 0.57 [6], 0.77 [13]
48%
37%
76%
0.5
0.3
0.77
30%
80%
$1,000
$5,000
$1,000
$3,000
probability of fall causing injury Probability of hip fracture from fall Probability of head injury from fall Probability of any fracture from fall Probability of wearing hip protector
Probability of hip fracture if wearing protector Reduction of fracture if fall on ‘safe floor’ Cost of hip fracture treatment Cost of head injury treatment Cost of other injury treatment
19.3%
33% [18], 63% [21], 80% 63% [22] $23,859 [24]
$23,859
$1,888 (i.e. €944) [8]
$1,888
$1,687.10 (derived from $1,687.10 8% of vertebral fractures getting hospitalized, and cost of other fractures i.e. .33*996 + .62*2191) [24]
8 of 17
Table 1: Epidemiological and cost parameters for base case and sensitivity analysis (cont.) Variable
Figures from literature
Base case
Cost of absorbent flooring per square metre Cost of hip protectors QALY loss in hip fracture
$150 [25]
$150
$163 first year, [26]
$163
Sensitivity Sensitivity (minimum) (maximum) $125 $175
0.2 [23, 24]
Experiments done on the new tiles indicate they are likely to last as long as the current tiles [18], which is equal to the lifetime of a building, approximately 40 years [2]. Assuming a residential facility with 50 residents, each with 20m2 space, the size of the residential facility is 1,000 sq. meters. This therefore requires an extra $75,000 more to replace the conventional flooring with the new tiles ($75*20*50). Spreading the value of the $75,000 spent in year one at a 5% interest rate over 40 years gives a total amount of $502,856, or $12,571 per annum, which translates to $251 per resident per annum.
2.6.3 Cost of providing hip protectors A New Zealand based study shows that each resident will need 3 hip protectors per year at a cost of $163 [26]. This comprises $160 for the protectors and $3 for program costs. It is assumed that the hip protectors are replaced every year. 2.7 Cost effectiveness and cost utility analysis The results compare the costs and outcomes of the current situation (no universal provision of hip protectors or use of injury reducing flooring) with the two strategies discussed above. Using the probabilities summarised in Table 1, the cost effectiveness analysis focuses upon the number of falls avoided. Incremental cost utility results are presenting using the QALY losses resulting from hip fractures. Sensitivity analysis identifies the robustness of the results to alternative values of the key parameters.
9 of 17
Table 2: Injuries and lost QALYs from falls
Current Hip protectors Absorbent flooring
Probabilities (Per person) Hip Any QALY Fracture Injury loss 1% 0.1325 0.002 0.0076 0.1301 0.00152 0.0037
0.049025
0.00074
Numbers (Per 10000 people) Hip Any QALY Fracture Injury losses 100 1325 20 76 1301 15.2 37
490
7.4
3. Results 3.1 Injuries from falls The expected number of hip fractures and total injuries expected for people in New Zealand was calculated using the probability of individuals sustaining an injury for each scenario. These terminal probabilities and numbers of injury per 10,000 individuals are shown in Table 2.
The results suggest that in the absence of any programme, there will be approximately 100 hip fractures for each 10,000 residents in residential facilities in New Zealand. The introduction of hip protectors will decrease the number of hip fractures by approximately 25%, while the absorbent floors might reduce hip fractures by 67% (as was assumed above).
Looking at the total number of injuries, there is expected to be approximately 1325 injuries in the absence of any programme. This reduces by only 2% were hip protectors to be introduced, reflecting the fact that hip protectors do not provide protection against all other types of fractures. However, the expected number of injuries would fall to only 490 should absorbent floors be introduced.
For each 10,000 people, the hip fractures could be expected to result in an annual loss of 20 QALYs per year. Introducing hip protectors reduces the QALY loss to 15.2 per year, while the absorbent flooring could be expected to reduce the QALY loss to 7.4 per year. 10 of 17
3.2 Cost The costs of the current situation and the two scenarios are shown in Table 3. Introducing hip protectors would be expected to result in an additional cost of $106 per person ($570 - $464), reflecting the additional cost of providing hip protectors but a per-person saving of $57 ($464 - $407) from the averted hip fractures. Absorbent flooring is expected to yield a cost savings of $41 per person ($464 - $423), due primarily to the reduced cost from both hip fractures and other injuries. Taken together, the introduction of hip protectors would expect to require an additional expense of over $1 million when compared to the current condition, while the absorbent flooring could be expected to result in a cost savings of $415,000 per annum.
3.3 Cost effectiveness and cost utility The results suggest that providing hip protectors will reduce hip fractures but at an additional cost. Compared with the current situation, the introduction of hip protectors will cost an additional $1,057,384 per annum (Table 4). There will be 24 fewer hip fractures (100 – 76), equating to a cost per hip fracture avoided of $44,058. Because other injuries are expected to be the same, this is also the cost per injury avoided. And hip protectors leads averts a loss of 4.8 QALYs per year, equating to a cost per QALY saved of over $220,000.
Table 3: Costs
Current Hip protectors Absorbent flooring
Costs (Per person) Prog. Injury Total $0 $464 $464 $163 $407 $570
Total costs (Per 10000 people) Prog. Injury Total $0 $4,643,453 $4,643,453 $1,630,000 $4,070,837 $5,700,837
$251
$2,510,000 $1,718,077 $4,228,077
$172
$423
11 of 17
Table 4: Cost differences per outcome per 10,000
Current vs. Hip Protectors Current vs. Absorbent Flooring Hip Protectors vs. Absorbent Flooring
Cost difference
Cost per hip fracture
Cost per injury
Cost per QALY loss
$1,057,384
$44,058
$44,058
$220,288
$415,375
($6,593)
($497)
($32,966)
$1,472,759
($37,763)
($1,816)
($188,815)
Impact absorbent flooring compares more favorably. The results suggest that impact absorbent flooring (strategy II) will result in cost savings, reduced hip fractures and reduced injuries. Compared to the current situation, strategy II saves $415,375 and averts 63 hip fractures, which equals $6,593 saved per averted fracture. In prevention of all injuries, strategy II saves $415,375 and prevents 835 injuries which is $497 saved per averted injury compared to the current situation. And it averts a loss of 12.6 QALYs at a cost savings of $32,966 per QALY saved. Thus, absorbent flooring dominates the outcomes from the current situation.
Absorbent flooring also dominates the introduction of hip fractures. For 10,000 elderly people, strategy II prevents 39 more hip fractures than strategy I and saves $1,472,759 which translates to $37,763 saved per averted hip fracture, $1,816 per injury averted and $188,815 per QALY averted.
3.4 Sensitivity analysis This model is quite sensitive to the efficacy of the new flooring in preventing fractures. If strategy II prevents only 30% of injuries, for every 10,000 residents, it costs $63,862 and prevents six hip fractures compared to strategy I, while it costs $1,121,246 and prevents 30 hip fractures compared to the current situation.
It is moderately sensitive to the risk of hip fracture upon a fall, with costs of $287,321 and savings of $1,056,371 compared to the do nothing approach and strategy I respectively if the risk of sustaining a hip fracture upon falling is 1%.
12 of 17
It is very sensitive to the cost of the new flooring: a $25 decrease in cost of the flooring triples the cost savings compared to the current approach, while a slight increase in flooring cost halves the cost savings compared to strategy I, and incurs rather than saving costs compared to the current situation.
It is minimally sensitive to the probability of falling, and very sensitive to the cost of treating head injuries, as these are the commonest type of injury sustained. Table 5 shows results of one way sensitivity analysis for the ICER (incremental cost effectiveness ratio).
Table 5: One way sensitivity analysis for 10,000 residents
Base case cost (savings) Flooring effectiveness 30% Flooring effectiveness 90% Flooring cost $125 per sq. meter Flooring cost $175 per sq. meter Hip fracture rate 1% Head injury cost $500 Head injury cost $5,000
Costs (savings) ($) Vs current Vs hip protector (411,093) (1,468,477) 1,121,246 63,862 (1,664,826) (2,722,210) (1,249,187) (2,306,571) 427,001 (680,383) 287,321 (1,056,371) 419,625 (637,759) (2,273,625) (3,331,009)
3.5 Breakeven analysis A breakeven analysis was carried out, to find out the level of effectiveness of the new flooring in preventing injury at which the cost of introducing this floor would be equal to the cost of a) the current approach and b) use of hip protectors. All other variables were held constant. The results suggest that the absorbent flooring would need to reduce the risk of injury upon falling by 31.2% to cost as much as the hip protector strategy, and by 54.1% to cost as much as the current system. The results are as shown on table 6. Table 6: Breakeven analysis Effectiveness of absorbent floor in preventing injury Cost per 10,000 residents at breakeven point
vs. current approach 54.1%
Vs. hip protector 31.2%
$4,643,453
$5,700,837
13 of 17
4. Discussion The results suggest that impact absorbent flooring is cost effective when compared with either the current situation or the use of external hip protectors. Under the assumptions used in the model, it would expect to result in a reduction in the number of hip fractures, other injuries and QALY losses while costing less overall. These results are robust to changes in many of the key parameters of the model.
The savings per various types of injuries averted, and per QALY gained are significant. Most interventions cost some amount per QALY gained or per clinical or other desirable outcome attained. That the new technology will both result in cost savings and better outcomes makes it very desirable. That it is not dependent on compliance by the aged residents ensures no variation in effectiveness between different institutions, regardless of the cognitive status of the residents.
Sensitivity analysis examined the robustness of the results to changes in key parameters, particularly the cost of the flooring and the rate of injuries. Overall, the results were shown to be fairly robust to changes in these assumptions. Various assumptions, however, were made in this model. These include: that the new flooring reduces the risk of all injuries at the same rate; that it will last as long as conventional flooring and will not increase the propensity of the elderly persons to fall; and that hip protectors do not increase the likelihood of a fall.
The model combined all other injuries except fractures and head injuries under other injuries. This potentially reduces the cost effectiveness of the new flooring technology. Thus, the results should be treated as a conservative estimate of the actual cost savings that are expected to result. Various studies indicate that several other injuries occur on falling, with soft tissue injuries occurring in 10-28% of fallers [4, 5]. Some of these need suturing, or other treatment hence incurring various costs.
14 of 17
The cost of treating head injuries varies greatly with severity of injury, and the amount used in this study is a mere estimate from an average cost of treating fall-related injuries among institutionalized elderly individuals in Finland [8]. This weakness has been countered by varying widely the cost of treating head injuries in the sensitivity analysis.
Another limitation of this study is that in the cost utility analysis, it considered only hip fractures. This is because of the potential difficulty associated with assigning utility values to injuries lumped together as ‘other fractures’ or to head injuries, as these have differing severities. Utility losses associated with hip fractures are available, hence only this outcome was considered. This therefore implies that the impact absorbent flooring actually results in more cost savings per QALY gained than estimated in this study.
5. Conclusion The impact absorbent flooring is more effective in reducing injuries upon a fall and less costly than the current approach as well as the use of hip protectors. It results in cost savings across a wide range of scenarios. The fact that it does not depend on compliance on the part of the individual adds to its intuitive appeal, and is likely to result in wide acceptance.
15 of 17
6. References [1] Sharma OP, Oswanski MF, Sharma V, Stringfellow K, Raj SS. An appraisal of trauma in the elderly. The American Surgeon. 2007;73(4):354-8. [2] Zacker C, Shea D. An economic evaluation of energy-absorbing flooring to prevent hip fractures. International Journal of Technology Assessment in Health Care. 1998;14(3):446-57. [3] Maki BE, Fernie GR. Impact attenuation of floor coverings in simulated falling accidents. Applied Ergonomics. 1990;21(2):107-14. [4] Gryfe CI, Amies A, Ashley MJ. A longitudinal study of falls in an elderly population: I. Incidence and morbidity. Age Ageing. 1977 November 1, 1977;6(4):20110. [5] Hill K, Schwarz J, Flicker L, Carroll S. Falls among healthy, communitydwelling, older women: A prospective study of frequency, circumstances, consequences and prediction accuracy. Australian and New Zealand Journal of Public Health. 1999;23(1):41-8. [6] Swerissen H, DenneKamp M. Effect of hip protectors in resideantial aged care settings. Australian Journal on Ageing. 2006;25(1):9-13. [7] Butler M, Kerse N, Todd M. Circumstances and consequences of falls in residential care: the New Zealand story. The New Zealand Medical Journal. 2004;117(1202):1. [8] Nurmi I, Luthje P. Incidence and costs of falls and fall injuries among elderly in institutional care. Scand J Prim Health Care. 2002;20:118-22. [9] Fischer ID, Krauss MJ, Dunagan WC, Birge S, Hitcho E, Johnson S, et al. Patterns and predictors of inpatient falls and fall-related injuries in a large academic hospital. Infection Control & Hospital Epidemiology [NLM - MEDLINE]. 2005;26(10):822. [10] Satin RW. Falls among older persons: a public health perspective. Annual Review of Public Health. 1992;13:489-508. [11] Forsen L, Sogaard AJ, Sandvig S, Schuller A, Roed U, Arstad C. Risk of hip fracture in protected and unprotected falls in nursing homes in Norway. Injury Prevention. 2004;10(1):16-20. [12] Kannus P, Parkarri J, Niemi S, Pesanen M, Palvanen M, Jarvinen M, et al. Prevention of hip fracture in elderly people with use of a hip protector. New England Journal of Medicine. 2000;343(21):1506-13. [13] Parker M, Gillespie WJ, Gillespie LD. Effectiveness of hip protectors for preventing hip fractures in elderly people: systematic review. BMJ. 2006 2 Mar 2006;332:571-4. [14] Kerse N, Peri K, Fitch J. Hip Protector Compliance Study. Wellington, New Zealand: Accident Compensation Corporation; 2006. [15] van Schoor NM, Asma G, Smit JH, Bouter LM, Lips P. The Amsterdam hip protector study: compliance and determinants of compliance. Osteoporosis International. 2003;14(4):353-9. [16] Nabhani F, Bamford JS. Impact properties of floor coverings and their role during simulated hip fractures. Journal of Materials Processing Technology. 2004;153-154:13944. 16 of 17
[17] Casalena JA. Modeling and anthropomorphic testing of a human hip impacting a dually stiff floor [Ph.D.]. United States -- Pennsylvania: The Pennsylvania State University; 1996. [18] ACMA. Kradal: product testing summary report. Upper Hutt, New Zealand: ACMA Industries; 2007. [19] Campbell AJ, Reinken J, Allan BC, Martinez GS. Falls in old age: a study of frequency and related clinical factors. Age Ageing. 1981 January 1, 1981;10(4):264-70. [20] Luukinen H, Koski K, Hiltunen L, Kivela S-L. Incidence rate of falls in an aged population in Northern Finland. Journal of Clinical Epidemiology. 1994;47(8):843-50. [21] Healey F. Does flooring type affect risk of injury in older in-patients? Nursing Times. 1994;90(27):40-1. [22] Simpson AHRW, Lamb S, Roberts PJ, Gardner TN, Evans JG. Does the type of flooring affect the risk of hip fracture? Age and Ageing. 2004;33:242-6. [23] Zethraeus N, Borgstrom F, Strom O, Kanis JA, Jonsson B. Cost-effectiveness of the treatment and prevention of osteoporosis--a review of the literature and a reference model. Osteoporosis International. 2007 Jan;18(1):9-23. [24] Brown P, McNeil R, Radwan E, Willingale J. The burden of osteoporosis in New Zealand: 2007-2020. Wellington: Osteoporosis New Zealand Inc.; 2007. [25] Bowmar J. Cost of Kradal flooring. Dunedin 2007. [26] Brown P, Willingale J, Abdu-Aguye S. Cost effectiveness of providing hip protectors to residential homes in New Zealand. Wellington: Accident Compensation Corporation; 2006.
17 of 17
