Reliability and validity of the medical outcomes study, a 36- item short-form health survey, (MOS SF-36) after one-year hospital discharge of hip fracture patient in a public hospital

Anan Udombhornprabha 1*, Jariya Boonhong 2,3, Tawechai Tejapongvorachai 4
  1. Program in Clinical Epidemiology, Faculty of Medicine, Chulalongkorn University, Rama 4 Road, Bangkok, Thailand
  2. Thai CERTC Consortium
  3. Department of Rehabilitation Medicine
  4. Department of Orthopedics, Faculty of Medicine, Chulalongkorn University, Rama 4 Road, Bangkok, Thailand
Corresponding Author: Anan Udombhornprabha, Program in Clinical Epidemiology, Faculty of Medicine, Chulalongkorn University, Rama 4 Road, Bangkok, Thailand,E-mail: [email protected]
Related article at Pubmed, Scholar Google
 

Abstract

Introduction: There is scarce of data in terms of health-related quality of life for hip fracture patients in Thailand due to the following: (i) lack of epidemiological aspects of hip fracture (eg. the relative incidence of osteoporosis, falls fractures and repeat fractures) in particular subgroups, (ii) lack of health status and quality of life aspects of both illness itself and the availability of different treatment options especially for elderly people. (iii) a substantial variation in terms of outcomes of care for patients and service for hip fractures. Objective: Hip fracture is a major healthcare burden in Thailand. This study explores quality of life for hip fracture patients from perspective of (i) Reliability of patient-reported outcomes (ii) Some clinical and demographic characteristics related to patient-reported outcomes

Method: Pre-hospital discharge 201 hip fracture patients were screened and follow-up over one year. Mail survey by a self-rated Medical Outcomes Study, a 36-item Short-Form Health Survey (Thai) dispatched for follow-up, other clinical and demographic characteristics were collected through direct interviews from patients or caregivers during recruitment with simultaneous crosschecking from medical records. A descriptive cross-sectional analysis was performed.

Result: Mails responder represented by 59.2% (N=119), with 36.1% (N=43) and 63.8% (N=76) for patient and caregiver rated outcomes. Mean(SD), [95% CI] score for physical, mental and global health of patient and care-giver rated outcomes of 36.2(10.6)[32.9-39.4], 54.5(10.0)[51.4- 57.5],43.9(9.3)[41.0-46.7], and 34.6(12.3)[31.7-37.4], 52.5(12.3)[49.6-55.3],42.7(11.1)[40.1- 45.2] were not statistically difference with p-value at 0.630,0.330 and 0.788 respectively. Respecting Cronbach’s alpha reliability coefficient by patients versus caregivers rated of the MOS SF-36 were 0.90 vs 0.91, 0.78 vs 0.84 and 0.90 vs 0.92. The presence of comorbidity significantly explains differences for quality of life outcomes in all health dimensions.

Conclusion: Health-related quality of life assessment with MOS SF-36 one year after hospital discharge for hip fractures patients was reliable. These results provided useful information related to post-treatment health-related quality of life outcomes. The shortcoming and limitation in terms of recall and report bias could be anticipated. The authors provided further justification for strength and weakness of implications for post-discharge health-related quality of life mail survey and future research aspects suggested.

Key words

Health-related quality of life, Medical Outcomes Study a 36-item Short-Form Health Survey (MOS SF36), Hip fracture, Reliability, Patient-reported outcomes.

Introduction

The rising incidence of hip fracture in public hospitals in Thailand cause huge impacts for increasing healthcare resource utilization (1,2). The population research in Thailand reported by John K et al (3) that older Thais with health as good or very good were increasing unless otherwise their basic activities of daily living on their own, but still had mobility problems due to lack of caregivers assisting them. In spite of this, hip fracture patients demand even more comprehensive care especially due to mobility difficulties. There is scarce of data in terms of health-related quality of life for hip fracture patients due to the following : (i) lack of an epidemiological aspects of hip fracture (eg. the relative incidence of osteoporosis, falls fractures and repeat fractures) in particular subgroups, (ii) lack of health status and quality of life aspects of both the illness itself and the availability of different treatment options especially for elderly people, (iii) a substantial variation in terms of outcomes of care for patients and service for hip fractures including those for prevention and management of risk factors beyond that explained by the difference in patients' characteristics, healthcare resources consumption, the differences in the severity, the types of fracture, specific socio-demographic and clinical characteristics of patients. This exploration initiated to understand the true outcomes valued by patients and information relevance to quality of life outcomes sought by patients and their caregivers after one-year post-discharge in general.

Objectives

To explore quality of life for hip fracture patients from perspective of (i) Reliability of patientreported outcomes (ii) Some clinical and demographic characteristics related to patient-reported outcomes

Material and Method

The Medical Outcomes Study, a 36-item, Short Form Health Survey, SF-36 developed, translated and validated into Thai by Jirarattanapholchai K et al (4) was obtained with permission to use for this study. The pragmatic setting in day-to-day clinical management of hip fractures started from patient admission in the emergency unit, hospitalized and subjected to surgical procedure as demanded by respecting orthopedist in charge, stabilized fracture for conventional or nonsurgical management for sufficient times until they should be discharged from hospital. All appropriate healthcare management strategies in terms of overall care and service were based on orthopedist in-charge pertaining to hospital budgeting and policy. These were essential elements for optimization of sustaining quality of service and care especially for public hospital in general. The study took place in January 2010. Hip fracture patients admitted in the hospital were both due to traumatic falls or repeated fracture of various aetio-pathologies. Admitted patients were either transferred patients from remote community regional hospital as well as from other provincial hospitals or patients from Chiang Rai Hospital. The study protocol was approved by Clinical Epidemiology Unit, Faculty of Medicine, Chulalongkorn University and by the research ethics committee of Chiang Rai Hospital with written informed consent obtained.

Sample, Sample size and Statistical Analysis

Hip fracture patients age 50 years or older admitted at the emergency unit were prospectively screened by attending orthopedists, as per eligibility criteria, including any diagnosis of hip fracture per ICD-10 except for hip fractures due to major traumatic accident such as car accident. All hip fractures were confirmed by positive radiography at the time of hospital admission. Before hospital discharge, research assistant nurse had assisted to ensure patients or caregivers properly understood the dialects among admitted patients should they speak northern dialects. Patients and patients’ caregivers were explained by research assistant nurse for the strict followup after one year post-discharge by a 36-item medical outcomes self-rated short-form health survey questionnaire (MOS SF-36). The procedure was to ensure correct understanding of the wording used in the health survey questionnaire which should have been mailed to them after one year post-discharge. Demographic and clinical characteristics of hip fractures were obtained both during patient interview before hospital discharged with simultaneous cross checking from medical records. Readmitted patients were recorded with periodic telephone monitoring. The MOS SF-36 (Thai version) with instruction mail in detail, together with pre-paid postage was dispatched to each of every 201 patients during March-May 2011 after one year post-discharged. The sample size, n as 95% CI of μ = ± 3, n = [(Z α//2 SD) / d]2, for d = 3 with α = 0.05 and standard deviation (SD) = 15 and Z α1/2 =1.96. The selected SD was based on an average SD of SD from a single mean score of SF-36 from Jithathai J et al (5) and from Charoencholvanich K et al (6), resulted in a sample size of 96 patients, assuming 75-80 % responder rate, then a sample size of 125 patients was needed. Statistical analysis included the reliability of each of health dimensions, health domains and global health, comparison of MOS SF-36 score rated by patients and caregivers and the impacts of demographic and clinical characteristics toward the health domains scores. The above were estimated using both internal consistency and item-scale correlation and inter-scale correlation. Internal consistency reliability was estimated with the Cronbach’s α and item-scale correlation and inter-scale correlation (7). The nonparametric statistical analysis for selected demographic and clinical characteristics with MOS SF-36 score was performed. All analyses were performed with the software SPSS version 16.0 (SPSS Inc.)

The Medical Outcomes Study, a 36-item Short-Form Health Survey Assessment

The Medical Outcomes Study, a 36-item, Short-Form Health Survey (SF-36) originally proposed by Ware JE et al (8) consists of 36 items and were combined to measure eight health symptom dimensions namely: Physical Functioning (PF), Role Limitations due to Physical Health (Role Physical, RP), Bodily Pain (BP), General Health Perceptions (GH), Vitality (VT), Social Functioning (SF), Role Limitations due to Emotional Problems (Role Emotional, RE) and Mental Health (MH). There is in addition a single-item measure of Health Transition (HT). The eight health symptom dimensions were grouped into two domains components of health domain, namely Physical Component Summary (PCS) and Mental Component Summary (MCS). The physical component summary domain consists of PF, RP, BP and GH whereas the mental component summary domain consists of VT, SF, RE and MH. The global health are scores of the overall 36-item combined (GLOBAL). The response choices for the items are on 2-, 3-, 5-, 6- point scales. The item scores range from 0 to 20,40,60,80,100 for 6-point scales and 0 to 25, 50, 75 and 100 for 5-point scales and in the same direction and as a reverted direction for specific items. The item scores with higher values indicate a better health states and less health limitation. The MOS SF-36 scales were scored using the method of standardized score from 0 to 100, for each of eight health symptom dimensions (PF, RP, BP, GH, VT, SF, RE and MH),each of two domains health component items (PCS, MCS) as well as for Global health score. Moreover, a recent three-component model for SF-36 score proposed by Suzukamo Y et al (9) namely physical component summary score (PCS) which consists of PF, BP and GH , a mental component summary score(MCS) which consists of VT, RE and MH and a role component summary score (RCS) which consists of RP, SF and RE. Only RCS was assessed. Statistical analysis with descriptive statistics were calculated for each characteristic of hip fracture patients both categorical variables and continuous variables including score for health dimensions, health domains and global health score.

Results

Demographic and clinical characteristics of hip fracture patients

Over all mails responders were 119 (59.2%). In spite of this low responder rate, authors had anticipated to recruit more patients as needed only 96 patients to have standard deviation of mean score of SF-36 closed to an estimate from previous study in Thai patients (5,6). Therefore, the overall hip fracture patients recruited was 201 patients which at least allowed a justification for adequate sample size. There were actual non-responders of 43 (21.3%) where mails returned with survey but did not complete the survey, whereas only 18 (8.9%) mails returned with no recipients. There were death reports of 21 (10.4%). This could justify for lower responder, as this was probably the untrue responder. Causes of death were mainly due to cardio-pulmonary collapse, systematic infection and renal failure which were consistent with subsequent complications after hip fracture both surgical patient or conventional treatment hip fracture patients. This was probably hip fracture patients in our analysis mostly were elderly patients with the median age was 79 years old. However, there was no specific registered death report from Chiang Rai hospitals where patients were admitted as genuine certified cause of death rather than death certificates from district hospitals records. The result was due basically to nature of national public hospital service system, especially for the tertiary hospital to obtain data transfer needed from referred patients. This issue remains key factor in assessing death with accurate mortality statistics in Thailand which could be anticipated elsewhere in Thai healthcare system as reported by Tangcharoensathien V et al (10). Male hip fracture represented slightly higher proportion (57.1%) than female (42.9%) which was similar to the finding from Rojanasathien S et al in the Northern Thailand (11). Whereas several other studies elsewhere, women was almost twice higher than men (12,14). Overall mean age of hip fracture patients was 74.7 year olds (SD±11.0), ranging from 50 - 104 year olds. Majority of patients were older than 65 years old which represented by 96 patients (80.7%). Majority of 111 (93.2%) patients were medically reimbursed through national universal healthcare insurance coverage (UC) which was a public healthcare scheme supported by Thai government. Main clinical characteristic were Intertrochanteric fracture (64.8%), and were managed by conventional or non-surgical management (59.6%). Overall, other demographic and clinical characteristics were given in Table 1.

Health symptom dimensions, domains, global health and reliability of score

There were 43 (36.1%) and 76 (63.8%) for patients and caregivers rated survey. Over all mean score, standard deviation and a 95% confidence interval between each of health dimensions, domains and global health score with their respecting Cronbach’s α coefficient for reliability assessment were given in Table 2. In comparing the different in scoring between patient-rated and caregiver-rated by a nonparametric statistical analysis, we found that there was no outstanding difference of outcomes score except for Role limiting emotion (RE) where patients tended to rate higher or better health for themselves (51.7 , 48.1-55.3) than their caregivers (43.3, 39.6- 46.9) with a significance at p = 0.011. The scores for each symptom dimensions were reliable for physical function, physical component summary score and global health score both by patient-rated and caregiver-rated (Cronbach’s α coefficient = 0.945 and 0.957, 0.911 and 0.918, 0.901 and 0.925 respectively). We found that the concepts of three symptoms domains proposed by Suzukamo Y et al (9) reflected health component closer to both physical function (PF) and role limiting physical (RP) and mental health (MH), where patients may feel indifference toward one another. Since, the role component summary score (RCS) though had exhibited an insight about their own ability to perform, where patients tended to rate higher score of role for themselves with mean (SD) [95% CI] score of 44.4 (9.4) [41.51-47.29] as compared with the caregivers 38.7(11.2)[36.14-41.26] (p=0.014), the RCS was given in Table 2. Though the RCS in our setting was less reliable with a Cronbach’s α coefficient for RCS between 0.57 for patients and 0.73 for caregivers, suggesting that the RCS was more cultural specific. Other scores mostly reflected high reliability, majority of which reflected Cronbach’s α coefficient ranged from 0.77 and above 0.91 for PCS, MCS and Global Health which were given in Table 2. The inter-scale correlations and internal consistency reliability provided in Table 3 revealed that the reliability Cronbach’s α coefficient or the within-scale correlation (diagonal bold) was higher than the between scale correlation which confirmed that each symptoms dimension for Thai hip fracture was generally being highly discriminated between different health symptom concepts measured except for physical function, role limiting physical, general health, vitality and role limiting emotion were less discriminated.

Factors contributing to quality of life for Thai hip fracture patients

The demographic and clinical characteristics taken for comparison were gender, age with a level of 65 years old, a type of medical insurance received, alcohol drinking and cigarette smoking status, pattern of fracture (femoral neck fracture and intertrochanteric fracture), type of hospital management of hip fracture (surgical and non-surgical), body mass index (BMI) (at level of 20 Kg/m2) and presence of comorbidity. The BMI at level of 20 Kg/m2 was used for comparison as earlier study by Laet C et al (15) had been concluded that the BMI at level of 20 Kg/m2 as compared with 25 Kg/m2 was associated with a nearly twofold increase risk ratio from RR of 0.83; 95%CI (0.69-0.99) to RR of 1.95; 95% CI (1.71-2.22) or over twofold.. However, the BMI (at cut-point level of 20 Kg/m2) in this study did not render evidence for significant different impacts on MOS SF-36 score. The BMI for hip fractures patients, in Thai patient setting in this study reflected mean (SD) BMI of 18.96 (2.57) Kg/m2. This value for hip fracture patient was even higher than the value for the underweight in general Thai suggested by the population health survey among a Thai cohort population health study of 38,815 men and 47,070 women reported by Lim LLY et al (16). However, patterns of hip fracture whether femoral neck (N=42, 35.3%) or intertrochanteric fracture (N=77, 64.7%) probably attributed to different health-related quality of life for all health dimensions including physical, mental and global health score. These reflected scores difference with p-value of less than 0.282, 0.119 and 0.188 respectively (Table 4.2). The body mass index cut-point level of 20 Kg/m2 did not reveal different in all health domain score (p=0.445, p=0.325 and p=0.693) (Table 4.2). The type of management between surgical and conventional or non-surgical revealed slightly different in mental health domain score (p=0.280) but not for physical and global health (p=0.665 and P=0.386) (Table 4.2). The presence of comorbidity suggested the outstanding impacts toward health-related quality of life which reflected on all health domain scores with significant difference for physical, mental and global score at p-value =0.003, <0.001, <0.001 respectively (Table 4.2). Whereas some demographic characteristics such as gender suggested some impacts on health-related quality of life which reflected some score difference only for mental domain score at p= 0.013 and slightly for global health with p=0.103. Women tended to have better physical health, mental health and global health than men for all domains (Table 4.1). Overall, the comparisons of MOS SF-36 score were given in Table 4.1. Even though alcohol drinking and cigarette smoking were detrimental risk factors for future fracture especially for hip fracture reported in a meta-analysis by Kanis JA et al(13,17), authors found that these characteristics suggested less impact which reflected no statistical significant difference in terms of patient-reported outcomes with p-value of 0.685, 0.847, 0.720 and 0.654, 0.562, 0.840 for physical, mental and global score respectively (Table 4.1).

Discussion

This exploration though exhibits some limitations as cross sectional mail survey. Since the study employed direct interview with pre-discharge education and consultation for further follow-up after one year post hospital discharge to patients and caregiver with specialized research nurse. The researcher-patient communication via local dialect ensured that the mails responding to survey after discharged was reflected from patients and caregivers true understanding even after discharge at home rehabilitation. The authors found that most patient-reported outcomes reflected by both patients and their caregivers were under-estimated with lower score. This was probably due to the healthcare system provision in the public hospital and is specific to universal healthcare system. This system allows all patients to have been provided free access to health service. Therefore, patients and their caregivers tended to be intimidated should their well-being status implicated in less service and care given by the healthcare provider, thus could possibly intriguing a report bias. The authors proposed that the mails survey with pre-counseling before hospital discharge should provide remedy to reported bias. Moreover, comparison of score rated by patients and caregivers were not significant different in most of score and sub-score indicating reliability of the MOS SF-3 from the mail survey. However, minor different for score of role emotion by patient which tended to be over-reported as compared with their caregiver was noted (Table 2). The authors proposed that recall bias persisted since the health-related quality of life at the time of hospital discharge and one year after hospital discharge could have rendered cognition bias. The bias may probably outstanding different especially for hip fracture patients treated by conventional treatment or non-surgical and surgical treatment and should have been explored specifically for further investigation. However, the attempt of this exploration was based on long-term consequence, health-related quality of life status perceived by patients and caregivers after home rehabilitation. In addition, it was envisaged that all hip fracture patients received standard medical treatment as provided through universal healthcare insurance by public healthcare provider. Thereby the mail survey could be justified with less recall bias, with provisional pre-discharge consultation. The low responder rate of only 59.2% was in fact adequate with sample size of 113, which higher than was originally estimated. It was thus sufficient to determine mean impact of the MOS SF-36 score measured. The reported mortality of 10.4% (N=21), probably underestimated, may possibly due to some reasons. Firstly there were mails return of 8.9% (N= 18) which could have otherwise being included as death. These mails may have been perceived by local responder caregivers to reject mails send to deceased patient. Secondly, the lower mortality was probably due to our sample had more non-surgical discharge of 59.6% (N=71) as compare with surgical discharge 40.4% (N=48). The one-year mortality reported was closed to earlier report among Thai hip fractures which reported as high as 17% by Chariyalertsak, S. et al (18 ) and nevertheless lower than a one year post-surgery mortality of 27.1 % for hip fracture reported by Lee AYJ et al (19). The cause of dead included cardiopulmonary collapse, systematic infection, probably as consequence of post-discharge surgical complications. In fact in our case, there was as high as 80.9% (N=17, 14 women and 3 men) post-surgical hip fracture patients for all total reported dead (N=21). There was only 19.1% (N=4) non-surgical patients as post-discharge mortality. Since overall surgical discharge was 40.4% (N=48) among all 119 cases. As such, overall after one-year post-discharge mortality for surgical management could be as high as 35.4% (17/21) and this was closed to finding reported earlier by Lee AYJ et al (19). The public hospital with high bed turnover rate may often override to expose risk especially for surgical hip fracture patients. The authors found that in this case, hip fracture patients admitted in the hospital was ranged from 4 to 21 days before discharge. The authors found that long hospitalization period and time to surgical were not relevance to hip fracture mortality. However, this needs further investigation. As this investigation conducted in a tertiary hospital receiving transferred patients, a long delay from the time of fracture to hospital admission may be main factor to mortality similar to that reported by Vidal, E. I. et al (20). The authors found that female hip fracture had higher post-discharge mortality with 71.4% (N=15), similar to the finding from by Gronskag, A. B et al and Yoon HK et al (21,22). Among overall hip fracture, low physical score from MOS SF-36 was noted even after one year hospital discharge with home rehabilitation. In spite of overall health limitation especially in terms of physical health, patients experienced improving health-related quality of life especially with better mental function which may be more substantive to overall global health score. Posthospital discharge with supportive home care, community service and proper rehabilitation were major factors contributing to better quality of life for hip fracture patients regardless of other factors as reflected in this exploration. Suriyawongpaisal, P et al (23) conducted a longitudinal analysis of hip fractures follow-up for 19 months, suggested that poor quality of life was the main cause of mortality. They concluded that comorbidities were factors contributing to overall poor health-related quality of life score assessed with SF-12. The authors found that, with MOS SF-36, comorbidities were essential factor contributing to overall poor health-related quality of life which could possibly leading to death with consistent with the finding reported by Suriyawongpaisal P et al (23). Moreover, mental health played important role in overall healthrelated quality of life. The supportive intervention, a so-called interdisciplinary intervention for older hip fracture patients after surgery suggested by Shyu YIL et al (24) had confirmed to improve health-related quality of life assessed with SF-36 as compared with patients who did not have interdisciplinary intervention. This possibly explains the improvement of mental health through interdisciplinary intervention. Unless otherwise, better management of pre-fracture in terms of prevention of falls leading to fracture, better medical surveillance to prevent any susceptible complications, urgent extensively management of comorbidities plays key role in overall healthcare. For a better prospective recruitment of hip fracture patients as multicenter research is encouraging to ensure better representative of younger age group to support for further understanding in terms of quality of life of after hip fracture. This exploration though had limitation and shortcoming as a cross-sectional by nature, the authors found that patient-reported outcomes provided deeper understanding of outcomes valued by patients. The MOS SF-36 was a good assessment tool with reliability to discriminate heath symptoms. The MOS SF-36 assessment is a good tool complimentary to clinical follow-up and management of hip fracture patients. The MOS SF-36 assessment was a reliable tool though may have some recall bias if the assessment being conducted very shortly after hospital discharge. The pre-discharge counseling and the extended assessment period just before hospital discharge and thereafter 6-12 month as follow-up could possibly reduce recall bias. There may be critics of report bias as some mental function from by caregiver-rated tended to rate better score than patient-rated. This could be minimized by appropriate orientation for both patients and caregivers to approach toward better home rehabilitation. The authors found that MOS SF-36 as supporting tools for both caregivers and patients could be employed along clinical outcomes assessment. This approach could help establish as early as possible before hospital discharge, the direction to which patients turn to health-related quality of life. Future research looking at health-related quality of life assessment with MOS SF-36 in a prospective approach among hip fracture patients with direct surgical and traditional management, within the context of healthcare resource utilization albeit the concurrent management of comorbidities could support for healthcare provider for proper resource allocation decision.

Conclusions

This study demonstrated that the Medical Outcomes Study, a 36-item Short-Form health survey is a reliable instrument complimenting clinical outcomes assessment for hip fracture. It is reliable whether patient-rated or caregiver-rated and the outcomes score could well discriminate healthrelated symptoms. The limitation of assessment could be offset by appropriate monitoring to help healthcare service officer to establish for better source to which patients turn for their health. Essentially, the presence of comorbidities as clinical risk factor of hip fracture plays a key role in health-related quality of life outcomes for hip fracture patients. Managing care and service during hospitalization, before and after hospital discharge especially comorbidities were essential elements to render good outcomes for subsequent home rehabilitation after one year postdischarge hip fracture patients.

Conflict of Interest

None declared.

Acknowledgement

We thank the orthopedics department of Chiang Rai hospital especially Dr. Pragun Sukwong M.D. for all kind supports in the registered and recruitment of patients. We also thanks the orthopedics ward house nurse chief nurse, Jamjit Thepnamwong M.S.N., RN for arranging external training for patients and patient's caregiver for a face-to-face interview, explanation of the mailing follow-up during their home rehabilitation by pre-paid mails before hospital discharge. We also thank for the Clinical Epidemiology unit of the Faculty of Medicine, Chulalongkorn University for partially support for our investigation.
image
image
image
image

References

1. Reungjarearnrung K, Chonglerttham S, Permsuwan U. Healthcare resource utilization and outcomes analysis of osteoporotic hip fractures among Thai Post-menopausal women: A Pilot Investigation. Value in Health 2010; 13 7:A558

2. Srisawai K, Thepnamwong J, Sukwong D.P., Reungjarearngrung K., Healthcare resource utilization of Thai hip fracture patients in public hospital: Cost of illness analysis at Chiangrai hospital. Value in Health 2011; 14 7:A562

3. Knodel J, Chayovan N. Gender and Ageing in Thailand: A Situation Analysis of Older Women and Men. Population Studies Center Research Reports.(2008), Population Studies Center, University of Michigan Institute for Social Research. Available online at https://www.psc.isr.umicu.edu/pubs/abs/5642.com/ Accessed February 12, 2012.

4. Jirarattanaphochai K, Jung S, Sumananont C, Saengnipanthkul S. Reliability of the medical outcomes study short-form survey version 2.0 (Thai version) for the evaluation of low back pain patients. J. Med. Assoc. Thai. 2005; 88 10: 1135-1138.

5. Jithathai J, Komsopapong L, Songjakkaew P, Kongsakon R. Health-related quality of life after hip fracture in the elderly community-dwelling. Southeast Asian J Trop Med Public Health. 2003; 34 3: 670- 674.

6. Charoencholvanich K, Pongcharoen B. Oxford Knee Score and SF-36: Translation and reliability for use with total knee arthroplasty patients in Thailand. J Med Assoc Thai. 2005; 88 9: 1194- 1202.

7. Chronbach L. Coefficient alpha and internal structure of the test. Psychometrika 1951; 16:297-304.

8. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). Med Care. 1992; 30 6: 473-83.

9. Suzukamo Y, Fukuhara S, Green J, Kosinski M, Gandek B, Ware JE. Validation testing of a threecomponent model of Short Form-36 scores. J Clin Epidemiol. 2010; 64 3:301-308.

10. Tangcharoensathien V, Faramnuayphol P, Teokul W, Bundhamcharoen K, Wiboolpholprasert S. A critical assessment of mortality statistics in Thailand: potential for improvements. Bull World Health Organ. 2006; 84 3: 233-238. Available online at https://www.scielosp.org/scielo.php?pid=S0042- 96862006000300019&script=sci_arttext&tlng=es. Accessed Feb 12, 2012

11. Rojanasathien S, Leuvitoonvejchkij S, Epidemiology of Hip Fracture in Chiangmai. J. Med. Assoc. Thai 2005; 88 5: s105-s109

12. Cumming RG, Nevitt MC, Cummnings SR, Epidemiology of Hip Fractures. Epidemiol Rev. 1997; 19 2: 244-257

13. Kanis JA, Johnell O, Oden A, Johansson H, De Laet C, Eisman JA, Fujiwara S, Kroger H, MaCloskey EV, Mellstrom D. et al. Smoking and fracture risk: a meta-analysis. Osteoporos Int. 2005; 16 2: 155-162.

14. Kannas P, Parkkari J, Sievanen H, Heinonen A,Vuori I, Jarvinen M. Epidemiology of hip fractures. Bone 1996; 18 1Suppl 1: S57-S63.

15. Laet C, Kanis J, Oden A et al. Body mass index as a predictor of fracture risk: A meta-analysis. Osteoporos Int. 2005; 16 11: 1330-1338.

16. Lim Lyntte LY, Seubsman S, Sleigh A. Validity of self reported weight, height, and body mass index among university students in Thailand: Implications for population studies of obesity in developing countries. Population Health Metrics 2009; 7 15: 1-8

17. Kanis JA, Johansson H, Johnell O, Oden A, De Laet C et al. Alcohol intake as a risk factor for fracture Osteoporosis Int. 2005; 16 7: 737-742.

18. Chariyalertsak S, Suriyawongpaisal P, Thakkinstain A. Mortality after hip fractures in Thailand. Int Orthop 2001; 25: 294-297.

19. Lee AYJ, Chua BSY, Howe TS. One-year outcome of hip fracture patients admitted to a Singapore hospital: quality of life post-treatment. Singapore Med. J. 2007; 48 11: 996-999.

20. Vidal, E.I., Moreira-Filho,D.C., Coeli, C.M., Camargo Jr., K.R., Fukushima, F.B., et al. Hip fracture in the elderly: does counting time from fracture to surgery or from hospital admission to surgery matter when studying in-hospital mortality? Osteoporos Int 2009; 20 5:723-729

21. Gronskag, A. B. Romundstad, P. Forsmo, S. Langhammer, A. Schei, B. Excess mortality after hip fracture among elderly women in Norway: The HUNT study. Osteoporos Int. 2012; 23 6:1807-1811

22.Yoon HK, Park C, Jang S, Lee YK and Ha YC. Incidence and Mortality Following Hip Fracture in Korea. J. Korean Med. Sci. 2011; 26 8:1087-1092.

23. Suriyawongpaisal P, Chariyalertsak S, Wanvarie S. Quality of life and functional status of patients with hip fracture in Thailand. Southeast Asian J Trop Med Public Health 2003; 34 2: 427-432.

24. Shyu YIL, Liang J,Wu CC,Cheng HS, Chen MC. An interdisciplinary intervention for older Taiwanese patients after surgery for hip fracture improves health-related quality of life. BMC Musculoskeletal Disorders 2010; 11:225 doi:10.1186/1471-2474-11-225
Select your language of interest to view the total content in your interested language

Viewing options

Post your comment

Share This Article

Flyer image
 

Post your comment

captcha   Reload  Can't read the image? click here to refresh