Abstract:
Purpose: Smoking is a risk factor for complications in orthopaedic surgery. The goal of this study was to explore whether there was discrimination against smokers in the referral process for isolated limb trauma.
Methods: A cross sectional study was undertaken on 166 consecutive ambulatory adult patients with isolated limb injury who were admitted to the orthopaedics service in a Level One Trauma Centre. Quality of care was assessed (analgesia, walking aids, immobilization) and delay for referral according to published expert recommendations.
Results: Among 166 consecutive patients referred, 45 were smokers. Smokers had lower family income (p=0.037), were younger (43 y.o. vs 50 y.o.; p=0.045) and used illegal drugs more often (16% vs 5%; p=0.045). Smokers were twice as likely to receive what was deemed “unacceptable” immobilization for their injury than non-smokers (52% vs 25%; p=0.002) and received inadequate walking aids (26% vs 9%; p=0.008). Delay from first primary care consultation to orthopaedic appointment was almost 2 times longer for smokers (93hrs vs 58hrs; p=0.034). Multivariable regression confirmed that smokers were more likely to receive unacceptable immobilization (adjusted OR: 3.6, 95% CI: 1.4 – 8.7) inadequate walking aids (adjusted OR: 3.0, 95% CI: 1.1 – 9.0) and had longer delays for consultation with an orthopedist. Confounding factors were no more significant in the model
. Discussion: Smokers who had a traumatic isolated limb injury received different care and had longer delays for orthopaedic consultations compared to nonsmokers. Whether this is due to medical bias or incorrect use of health services by patients remains to be determined.
Keywords:
Quality of care; Access to care; smoking; limb injury; fracture.
J.Orthopaedics 2012;9(4)e4
Introduction :
In orthopaedic surgery, smoking has a bad reputation. Moller reported a complication rate three times higher in smokers, and smoking was the strongest factor associated with complications after orthopaedic surgery [1]. From a biological perspective, it has been shown that bone healing is slower in mice exposed to cigarette smoke [2,3]. There are several other patient-related factors known to influence surgical outcome, such as age and comorbidities [4-7,]. Delays to receive appropriate care may also influence outcomes [8,9,] and in the case of traumatic orthopaedic injury, prompt access is important to prevent potentially serious complications[9-11].
Although the Canadian system is based on universal access to healthcare, timely access to a specialist may be problematic. The reasons for this may be due to the patient (not using health services appropriately), the system (long waiting lists, lack of proper triage), and the physician (favoring certain types of patients over others). The objectives of the present study were to explore whether patients who smoked received the same care as those who did not, in terms of quality of care and timeliness of care.
Material and methods
Study design
We conducted a cross-sectional study of 166 patients with isolated limb injuries, who were referred to orthopaedic services in a Level One Trauma Center. Inclusion criteria were: ambulatory adult patients, 18 years and over, with an isolated limb injury referred to the orthopaedic service. Exclusion criteria were: inability to speak French or English, and having an injury that occurred more than three months earlier. Following the first orthopaedic visit for an injury, patients meeting the inclusion criteria were recruited by a research nurse who administered a structured interview. Patients first answered a questionnaire that addressed socio-demographic data, injury history, past medical history, primary care, initial treatment, time of injury and time of each medical consultation.
A second questionnaire was completed by the treating orthopaedic surgeon regarding the type of injury, quality of immobilization, analgesia, walking aids, and adequacy of referral diagnosis as well as the type of treatment offered. A third questionnaire was completed by the patient regarding pain relief treatment, immobilization comfort, walking aids, and a pain evaluation, using a scale of 0 to 10 for pain level[12]. We set an acceptable level of pain to be 4/10, in accordance with the American Pain Society recommendations for acute pain management [13].
Measures
The independent variable, smoking, was assessed by the following question to the patient: “Do you currently smoke?” The number of cigarettes/day was also recorded as well as the duration of the smoking habit in years.
The dependent variables of access to care and quality of care were defined as follows. Access was measured by the delay in hours between seeing the first primary care doctor and the orthopaedist, as well as by the number of previous doctor’s visits prior to consultation with the orthopaedic surgeon. Quality of care was described in terms of analgesia, immobilization, walking aids and proper diagnosis. Pain level was determined when the patient saw the orthopaedic surgeon who classified the analgesia prescription.
Appropriate analgesia was determined by the patient if:
a) a prescription was given and considered sufficient in terms of pain relief, or
b) when the prescription was absent and the patient declared that he/she did not need pain killers.
Immobilization was defined as inappropriate if:
a) the joints proximal and distal to the injury site were not immobilized,
b) skin padding was absent,
c) inappropriate material was used,
d) the joint was immobilized in the wrong position, or e) there was a constricting bandage. This last criterion was assessed by the surgeon in terms of the patient’s comfort or by the presence of signs of compression. Walking aids was characterized as “inadequate” in the absence of a prescription for a patient who needed crutches, a cane, a wheelchair or a walker for lower limb injury. “Inadequate referral diagnosis” was defined as an absent or wrong diagnosis on the referral requisition.
Analysis
Descriptive data were used to characterize smokers and non-smokers in terms of age, sex, health status, types of lesion, time of injury, ethnic characteristics, and socio-economic status. All distributions of continuous variables were tested for normality.
Proportions were used to report categorical variables. We sought to identify the impact of smoking on the quality of care and access. We compared smokers and non-smokers in terms of quality and access using bivariate statistics.
We also used multiple regression models to determine whether smoking was associated with quality of care and access, adjusting for potential confounding factors such as: sex, age, education, legal status, family income, working status, financial compensation for the injury, country of birth, health status, co-morbidities, prescribed medications, alcohol and drug use. We included variables that had a p value <0.20 in the bivariate analysis as candidate variables in the multivariable models.
Results:
A total of 201 ambulatory patients were referred to the orthopaedic clinic for an isolated limb injury during the 4-month study period from September 1, 2006 to December 31, 2006. Of 178 eligible patients, 166 accepted to complete all questionnaires, for a response rate of 93.3%. The mean age was 48 years (SD 19; range 18-88) and there were 85 women (51%). The injuries are described in Table 1.
In this study 27 % (45/166) of the patients were smokers.
The smokers’ group was younger, used more illegal drugs and had a lower annual income (Table 2). Bivariate statistics indicated that smokers experienced longer delays to see an orthopaedic surgeon (93 hours for smokers vs. 58 hours for non-smokers). In terms of quality of care, smokers had twice as many inadequate or absent immobilizations and three times more absent walking aids in the presence of significant lower limb injury. (Table 2)
Regression models were constructed to determine whether smoking (adjusted for covariates) was associated with
a) longer delays to see the orthopaedist;
b) inadequate analgesia;
c) inappropriate immobilization;
d) no walking aids for those who needed one; d) inappropriate referral. These are described in Table 3 and indicate that smoking was significantly associated with inappropriate immobilization, inadequate walking aids and increased delay for consultation with an orthopaedist.
Discussion:
It seems that smokers with an isolated limb injury followed a different care and referral process compared to non-smokers. Our statistical analysis reveals that socioeconomic factors were not related to the differences observed between smokers and non-smokers. Smokers received suboptimal care in terms of walking aid prescription and quality of the immobilization following the injury [14]. Furthermore, delay from first primary care consultation to orthopaedic appointment was almost 2 times longer for smokers [15]. However, once referred, smokers were more likely to have appropriate referral diagnoses.
Differences between smokers and nonsmokers regarding care of post traumatic limb injury, but prior to orthopaedic consultation, have never been reported in the literature. Doctors’ perceptions of smokers may be different than nonsmokers, as one study indicated that doctors linked smoking to inter personnel problems and personality pathologies[16]. Although smokers and non-smokers appear to have equivalent access to some specialized cardiology services [17], there are indications of bias by physicians against smokers [18-19]. Smoking does appear to be associated with inferior surgical outcomes such as wound healing, complications (infections, delayed union) and function [20-,22] . Biological hypotheses have been used to explain these poorer outcomes among smokers [2,3].
However, health care practitioners and patient behavior can also lead to lower quality of care and outcome [23]. The relation between health care usage and variation in outcome has already been reported to explain the lower general health of smokers and increased mortality [24-25]. From a philosophical point of view, some justify discriminating against smokers using the following argument. Smokers cause themselves to be ill and therefore use more healthcare than nonsmokers. In an environment of limited healthcare budgets, this may decrease health care availability for non-smokers. Wilkinson counters that this argument is not cogent [26] . Smokers die younger and this decreased life expectancy compensates (in terms of health care resource use) for higher need during their life.
In other words, nonsmokers will end up consuming as much or more healthcare resources since they live longer and will require health care as they age. Thus, even in the context of limited resources, he believes that the equity between smokers and non-smokers should be maintained.
There are several limitations to our study. We relied on patient’s descriptions of events and previous primary care visits. The information written on the orthopaedic consultation requisitions were minimal in most cases. We hoped to reduce recall bias by limiting the time from the injury to 3 months. At the same time, by excluding cases referred more than 3 months post injury, we may have introduced a selection bias. Also, we evaluated only patients referred to the orthopaedic surgeon.
The primary care physician may have referred some patients to other specialists such as; physiatrists, plastic surgeons, sports medicine doctors, rheumatologists, or physiotherapists. Finally, we don’t know if the primary care physician was aware of the smoking habit of the patient.
In conclusion, injured smokers received different care and had longer delays for orthopaedic consultations.
Primary care quality and delay for orthopaedic consultation were associated with smoking status, possibly due to medical bias or incorrect use of health services by patients. Smoking is a known risk factor for complications in orthopaedic surgery. It is possible that inferior surgical outcomes may be related to different peri-operative care offered or used by smokers.
Acknowledgements
Dr. Rouleau was supported by a fellowship from « Fonds de Recherche en Santé du Québec », and from Mentor-CIHR (Canadian Institut Health Research)
Dr. Feldman was supported by a career award from the Arthritis Society and currently holds an award from the Fonds de Recherche en Santé du Québec
Dr Parent was supported by the « Fonds de Recherche en Santé du Québec » award.
The authors declare that they have no conflict of interest.
Reference:
1.Moller AM, Pedersen T, Villebro N, et al (2003) Effect of smoking on early complications after elective orthopaedic surgery. J Bone Joint Surg
85(2):178-81.
2.El-Zawawy HB, Gill CS, Wright RW, et al (2006) Smoking delays chondrogenesis in a mouse model of closed tibial fracture healing. J Orthop Res 24(12):2150-8
3. Skott M, Andreassen TT, Ulrich-Vinther M, et al (2006) Tobacco extract but not nicotine impairs the mechanical strength of fracture healing in rats. J Orthop Res 24(7):1472-9
4.Guo JJ, Yang H, Xu Y, et al (2009) Results after immediate operations of closed ankle fractures in patients with preoperatively neglected type 2 diabetes. Injury 40(8):894-6
5.
Pull ter Gunne AF, Cohen DB (2009) Incidence, prevalence, and analysis of risk factors for surgical site infection following adult spinal surgery. Spine 34(13):1422-
6.Shebubakar L, Hutagalung E, Sapardan S, et al (2009) Effects of older age and multiple comorbidities on functional outcome after partial hip replacement surgery for hip fractures. Acta Med Indones 41(4):195-
7.SooHoo NF, Krenek L, Eagan MJ, et al (2009) Complication rates following open reduction and internal fixation of ankle fractures. J Bone Joint Surg Am 91(5):1042-\
8. Amin A, Bernard J, Nadarajah R, et al (2005) Spinal injuries admitted to a specialist centre over a 5-year period: a study to evaluate delayed admission. Spinal Cord 43(7):434-7
9. Payne R, Kinmont JC, Moalypour SM (2004) Initial management of closed fracture-dislocations of the ankle. Ann R Coll Surg Engl 86(3):177-81.
10. Tripuraneni K, Ganga S, Quinn R, et al (2008) The effect of time delay to surgical debridement of open tibia shaft fractures on infection rate. Orthopedics 31(12)
11. Khatod M, Botte MJ, Hoyt DB, et al (2003) Outcomes in open tibia fractures: relationship between delay in treatment and infection. J Trauma 55(5):949-54
12. Stalnikowicz R, Mahamid R, Kaspi S, et al (2005) Undertreatment of acute pain in the emergency department: a challenge. Int J Qual Health Care 17(2):173-6
13. Dahl JL (2004) Pain: impediments and suggestions for solutions. J Natl Cancer Inst Monogr 32:124-6.
14. Rouleau DM, Parent S, Feldman DE(2010) Evaluation of primary care management for isolated limb injury: Study on 166 consecutive patients referred to orthopaedic surgery in a Level 1 Trauma Center. Clin Invest Med 33(2):E99
15. Rouleau DM, Parent S, Feldman DE (2009) Delay to Orthopedic Consultation for Isolated Limb Injury: A Cross Sectional Study in a Level One Trauma Center. Can Fam Physician 55(10):1006-7.e1-5
16. Harvey EL, Hill AJ (2001) Health professionals' views of overweight people and smokers. Int J Obes Relat Metab Disord 25(8):1253-61
17. Cornuz J, Faris PD, Galbraith PD, et al (2005) Absence of bias against smokers in access to coronary revascularization after cardiac catheterization. Int J Qual Health Care 17(1):37-42
18. Craig BM, Kraus CK, Chewning BA, et al (2008) Quality of care for older adults with chronic obstructive pulmonary disease and asthma based on comparisons to practice guidelines and smoking status. BMC Health Services Research8:144. doi:10.1186/1472-6963-8-144
19. Cyboran J (2005) Do you discriminate against smokers? Physicians earn failing grades on COPD report card. NRM 2(21): http://www.nationalreviewofmedicine.com/issue/2005/12_15/2_patients_practice02_21.html. Accessed 3 February 2010
20. Dinah AF, Vickers RH (2007) Smoking increases failure rate of operation for established non-union of the scaphoid bone. Int Orthop 31:503–505
21. McKee MD, DiPasquale DJ, Wild LM, et al (2003) The effect of smoking on clinical outcome and complication rates following Ilizarov reconstruction. J Orthop Trauma 17(10):663-7
22. Wang C, Li Y, Huang L, et al(2010) Comparison of two-staged ORIF and limited internal fixation with external fixator for closed tibial plafond fractures. Arch Orthop Trauma Surg 130 (10):1289-97 [Epub ahead of print]
23. Mooney GH (1983) Equity in health care: confronting the confusion. Eff Health Care 1(4):179-85.
24. Jha P, Peto R, Zatonski W, et al (2006) Social inequalities in male mortality, and in male mortality from smoking: indirect estimation from national death rates in England and Wales, Poland, and North America. Lancet 29; 368(9533):367-70
25. Jarvis MJ, Wardle J (2006) Social patterning of health behaviours: the case of cigarette smoking. In: Marmot M, Wilkinson RG (eds) Social determinants of health, 2nd edn. Oxford, Oxford University Press, pp 224–37
26. Wilkinson S (1999) Smokers' Rights to Health Care: Why the `Restoration Argument' is a Moralising Wolf in a Liberal Sheep's Clothing. J Appl Philos 16(3):255-69 Tables
Table 1. Injury description (Rouleau DM, Parent S, Feldman DE (2009) Delay to Orthopedic Consultation for Isolated Limb Injury: A Cross Sectional Study in a Level One Trauma Center. Can Fam Physician 55(10):1006-7.e1-5. printed with permission from Canadian Family Physician)
Characteristics |
Number |
Proportion (%) |
Injury type |
|
|
Fracture |
142 |
85.5 |
Soft Tissue |
24 |
14.5 |
Injury site |
|
|
Lower limb |
59 |
36 |
Upper limb
Perception of Severity
1 “not severe”
2 “partially severe”
3 “severe”
4 “very severe”
5 “extremely severe”
Most frequent fracture site |
107
6
74
53
14
15 |
64
3.7
45.7
32.7
8.6
9.3
|
Distal radius/ulna |
33 |
19.8 |
Metacarpal/hand phalanx |
28 |
18.5 |
Ankle |
19 |
11.1 |
Metatarsal/foot phalanx |
15 |
9 |
Most frequent soft tissue site |
|
|
Knee ligament |
5 |
3 |
Wrist sprain |
4 |
2.4 |
Elbow dislocation |
4 |
2.4 |
Ankle sprain |
3 |
1.8 |
Table 2. Differences between smokers and non-smokers with an isolated limb injury
Factors |
Smokers |
Non-smokers |
P value |
Age‡ |
43 years |
50 years |
p = 0.045 |
Annual family income < 30,000/year† |
44 % |
27 % |
p = 0.037 |
Illicit Drug Use† |
16 % |
5 % |
p = 0.045 |
Proper referral diagnosis† |
78 % |
58 % |
p = 0.019 |
Inadequate or absent immobilization† |
52 % |
25 % |
p = 0.002 |
Inadequate walking aid prescription† |
26 % |
9 % |
p = 0.008 |
Delays between 1st doctor and orthopeadic consultation‡ |
93 hours |
58 hours |
p = 0.034 |
† Chi square test
‡ Student T test
Table 3. Regression models of factors affecting Care in isolated limb injury.
Factors related to care |
WALKING AID†
Logistic regression |
Coefficient |
CI 95 % |
Smoking |
3.0 |
1.1 - 9.0 |
Unjustified referrals |
8.0 |
2.3 - 28.0 |
Immobilization |
3.6 |
1.2 - 10.7 |
IMMOBILISATION ‡
Logistic regression |
Coefficient |
CI 95 % |
Smoking |
3.6 |
1.4 - 8.7 |
Consultation to another center |
3.6 |
1.6 - 8.0 |
Alcohol |
0.3 |
0.1 - 0.7 |
Analgesia£
Logistic regression |
Coefficient |
CI 95 % |
Inappropriate walking aids |
3.7 |
1.4 – 9.9 |
Delay ¥
Linear regression |
Beta |
CI 95 % |
Smoking |
34.1 |
4.9 - 63.3 |
Multiple visits primary care |
89.3 |
56.1 – 122.5 |
Poor severity according to patient |
52.0 |
25.0 – 79.0 |
Consult to another center |
49.8 |
17.2 – 82.4 |
Lower limb injury |
39.5 |
11.3 – 67.7 |
† Walking aids quality has been used as a dichotomous variable: acceptable versus not acceptable.
‡ Immobilization quality has been used as a dichotomous variable: acceptable versus not acceptable.
£ Analgesia quality has been used as a dichotomous variable: acceptable versus not acceptable.
¥ Linear regression model showing the impact of factors related to an increased delay of referral (R2=0.277).
|
