HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Sidney Chocron Joseph Philippe Etievent Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Falcoz, P. E. Articles by Etievent, J. P. Search for Related Content PubMed PubMed Citation Articles by Falcoz, P. E. Articles by Etievent, J. P. Related Collections Cardiac - other Related Article

Ann Thorac Surg 2002;73:1222-1228
© 2002 The Society of Thoracic Surgeons

---

Original article: cardiovascular

Comparison of the Nottingham Health Profile and the 36-item health survey questionnaires in cardiac surgery

^a Department of Thoracic and Cardiovascular Surgery, Hôpital Jean-Minjoz, Besançon, France
^b Department of Biostatistics and Epidemiology, Faculté de Médecine et de Pharmacie, Besançon, France

Accepted for publication December 17, 2001.

* Address reprint requests to Dr Falcoz, Department of Thoracic and Cardiovascular Surgery, Hôpital Jean-Minjoz, Boulevard Fleming, 25000 Besançon, France
e-mail: pierre-emmanuel.falcoz{at}wanadoo.fr

	Abstract

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Background. Quality of life (QOL) instruments help to integrate the patient’s view into clinical practice and into the evaluation of new therapeutic strategies. The aim of the present study was to determine which of two generic QOL instruments, the Nottingham Health Profile (NHP) or the Short Form Health Survey (SF36), was the more suitable for use in cardiac surgery.

Methods. The NHP and the SF36 were compared before and 5 weeks after surgery. Comparison was conducted in two stages: (1) the acceptability and psychometric properties of the tools were measured, and (2) the short-time evolution of angina pectoris and dyspnea status were assessed with the QOL.

Results. A total of 322 patients were included and 299 patients completed preoperative and postoperative questionnaires. Acceptability was similar for both questionnaires. Internal consistency, ceiling effect, sensitivity to change, as well as the assessment of the evolution of angina pectoris and dyspnea were better for the SF36 than for the NHP.

Conclusions. The SF36 seems more suitable than the NHP for evaluating QOL in cardiac surgery.

	Introduction

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
In the field of cardiac surgery, results are usually evaluated in terms of mortality and morbidity rates. These criteria restrict the field of evaluation to cardiac symptoms. Certain side effects of surgery such as difficulty sleeping or limited physical mobility, which appear or worsen after surgery, affect the quality of life (QOL) of the patients and are not taken into account in traditional evaluations of surgical results.

One way of assessing the personal and social context of patients is to use QOL measures [1]. QOL instruments may be specific for a particular disease or group of patients or generic for all aspects of health-related quality of life (HRQL). As there is no specific questionnaire in cardiac surgery, we required candidate generic instruments to be self-administered, valid, available in French, concise, and previously applied in a sample of the general population. The two most commonly used questionnaires for evaluating QOL in cardiac surgery are two generic instruments: the Nottingham Health Profile (NHP) [2] and the Short Form Health Survey Questionnaire (SF36) [3].

The NHP has already been applied in cardiac surgery [4], particularly to compare preoperative and postopera-tive QOL [5]. The SF36 has been used in several studies in cardiac surgery [6, 7]. We felt it would be of interest to compare these two QOL instruments in order to know which one is more suitable for use in cardiac surgery. Selection of the most appropriate instrument depends on numerous factors, including measurement properties. Guidelines concerning the choice of instruments used to assess HRQL agree about the importance of acceptability and psychometric properties such as validity and sensitivity to change [8, 9].

The aim of this prospective study based on the completion of two self-administered QOL instruments, the NHP and the SF36, proposed before and 5 weeks after surgery, was to compare the measurement properties of these two questionnaires and the assessment of the short-term evolution of angina pectoris and dyspnea in this sample of patients.

	Material and methods

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Sample and questionnaires
This study was approved by our Institutional Review Board. From July 2000 to July 2001, 523 patients underwent open-heart surgery in the Department of Thoracic and Cardiovascular Surgery at the University Hospital of Besançon, France.

The two questionnaires were proposed to the patients by a data manager the day before open-heart operation (time 1 [T1]). Patients were also given self-administered questionnaires about angina pectoris and dyspnea. Those who were not fluent in French or who required unscheduled operations were excluded. 5 weeks later (time 2 [T2]), the patients who had answered the preoperative questionnaires were contacted by mail. They were sent a cover letter, the two QOL questionnaires, the self-administered questionnaires about angina pectoris and dyspnea, a questionnaire concerning their preference between the two QOL tools, and a stamped self-addressed return envelope. The order of administration of both questionnaires (NHP and SF36) was randomized for each patient and for each time of completion.

The NHP questionnaire is a widely used generic tool, originally written in English [2] and validated in French [10]. It contains 38 subjective statements divided into six sections: energy, physical mobility, emotional reactions, pain, sleep and social isolation. There are two possible responses per item: yes or no. Scores for each dimension range from 0 (normal health) to 100 (very poor health) and are calculated with weights determined by Thurstone’s method [11]. Its translation, which has been validated as correct in several languages, allows comparison of different cultures and populations.

The SF36 is a self-administered 36-item tool covering eight dimensions of health, including limitations in physical functioning, usual role activities, social functioning related to health problems and vitality. It also includes a global evaluation of health. Each dimension is scored on a scale from 0 to 100, with higher scores indicating better health. The number of possible responses per item varies from 2 to 6. The SF36 has received wide validation in English [12, 13]. The French version used here was adapted by forward and backward translation, iterative revision, and consensus by experts [14].

To facilitate comparison we normalized the dimensions of these two instruments using linear transformation to recode scores from 0 (poor health) to 100 (perfect health).

The assessment of angina pectoris and dyspnea, done by self-administered questionnaires given the day before open-heart surgery and 5 weeks afterwards, was considered as valid because the agreement between the coding of the patient and the medical coding (New York Heart Association [NYHA] and Canadian classification) were judged very satisfactory (kappa = 0.935 for angina pectoris and kappa = 0.879 for dyspnea).

The variables recorded were as follows: sociodemographic (age, sex, family situation, level of study), heart disease, angina pectoris status according to the Canadian classification, dyspnea class according to the NYHA classification, ejection fraction, comorbid diseases (diabetes mellitus, cerebral or peripheral vascular disease, renal failure, chronic obstructive pulmonary disease, obesity, depression and previous heart operation), surgical procedure, and postoperative complications.

Statistical analysis
Taking into account the standard deviation of the NHP [15] and the SF36 [6] scores, we assessed 350 to be the number of patients necessary for a relative error of 10% in measuring the score of the different dimensions.

Acceptability was assessed by comparing responses to the preference questionnaire using Mac Nemar’s ². The rate of missing items (completion rate) was assessed for each dimension and for each moment of completion. The total percentage of missing items (before and after surgery) of the two questionnaires was tested by the Fisher’s exact test. The time of completion for each questionnaire was tested by a paired comparison t test.

The "order" effect was tested by using a t test procedure on the means of the QOL score of both groups of patients (having received the NHP either in first or in second position) at T1.

Cronbach’s coefficient was used to calculate the internal consistency in all patients for the two questionnaires at T1. Values above 0.80 are considered to provide good internal consistency reliability, while those above 0.70 are considered adequate for short scales.

Floor and ceiling effects were determined by assessing the percentages of subjects with a score of 0 (poor health) or 100 (perfect health) for each dimension of each questionnaire at T1 and T2.

Concurrent validity was tested at T1 by assessing correlations between the NHP and the SF36 dimensions within a multitrait, multimethod matrix. Criterion validity was tested by assessing the correlations between the NHP and the SF36 dimensions and multiple disease-specific variables recorded preoperatively.

To assess sensitivity to change the patients were divided into three groups: "improved" when both angina pectoris and dyspnea were improved, "worsened" when either angina pectoris or dyspnea were worsened, and "unchanged" for the other patients. For each of these groups, we determined the standardized response mean (SRM) for each dimension of each questionnaire [16, 17]. A SRM value of more than 0.8 reflected an important change; between 0.5 and 0.8, a moderate change; and between 0.2 and 0.5, a weak change [18]. Analysis of variance adjusted for age and sex was used to compare the different scores of QOL, dimension by dimension, in the three groups.

A discriminant analysis was performed to find the most predictive dimensions of the patient’s status concerning their evolution in terms of angina pectoris and dyspnea (worsened, unchanged and improved) for each questionnaire. Variables with a level of significance less than or equal to 0.10 in the univariate analysis were included in the multivariate model.

For the NHP, the score of a dimension was not calculated if any item was left out [19]. For the SF36, we applied the scoring rules [20].

All tests were two-sided and due to multiple comparisons, only results with a p value of less than 0.01 were considered as statistically significant. All statistical analyses were performed with a statistical analysis system (SAS software, version 8.1; SAS Institute, Cary, NC).

	Results

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Characteristics of the study population
In all, 201 (38.5%) of the 523 operated on patients did not fill in the preoperative questionnaires: 12 (2.3%) did not speak French, 5 (1%) refused, and 184 (35.2%) required unscheduled operation. Thus, 322 completed preoperative questionnaires were available. At 5 weeks, 23 (7.1%) patients did not fill in the questionnaires: 9 (2.8%) patients died between the two times of completion and 14 (4.3%) did not answer despite numerous phone calls. Among the 299 patients who completed preoperative and postoperative questionnaires, 271 (90.6%) answered spontaneously and 28 (9.4%) answered only after numerous calls. For these 28 patients the postoperative questionnaire was obtained more than 8 weeks after their operation. Nevertheless their characteristics were similar to those of the patients who answered spontaneously (data not shown).

In terms of age-sex characteristics, the sample was approximately 70% male and 30% female, aged between 14 to 87 years (mean 65.7; SD 11; Table 1); 82% of the sample were married or cohabited and 53% lived in a rural area. More than half were white-collar (30%) or blue-collar (40%) workers. At the time of surgery, 274 patients (85%) were retired. The predominant heart valve disease was calcified aortic stenosis (37%).

At T1, 281 (87%) questionnaires for the NHP and 286 (89%) for the SF36 were completed with no missing items (p = 0.54). At T2, 261 (87%) questionnaires for the NHP and 286 (77%) for the SF36 were completed with no missing items (p = 0.001).

Fifty-three (18%) patients preferred the NHP, 34 (11%) preferred the SF36 (p = 0.042), and 212 (71%) expressed no preference. Preference was not influenced by gender (p = 0.16) but men found the iterative completion of the two questionnaires less tedious than women did (p = 0.0013).

The mean time for completion was 13.1 ± 9.5 minutes for the NHP, and 13.9 ± 9.7 minutes for the SF36 (p = 0.0012).

Psychometric properties of the instruments
Score distributions and internal consistencies
No significant difference in the results was caused by the order in which the questionnaires were proposed.

Table 3 shows the descriptive statistics of the two instruments at T1 and T2. At T1 the mean SF36 score ranged from 25.2 for the energy dimension to 72.0 for the social functioning dimension whereas the mean NHP score ranged from 63.6 for the energy dimension to 91.3 for the social isolation dimension. At T2 the mean SF36 score ranged from 15.9 for the physical role dimension to 68.2 for the social functioning dimension whereas the mean NHP score ranged from 62.7 for the energy dimension to 91.2 for the social isolation dimension.

View this table: [in this window] [in a new window]   Table 3. Features of Score Distributions (Linear Transformation to Uniform Scales Ranging Between 0 and 100), Internal Consistency (Chronbach’s ) of the Nottingham Health Profile (NHP) and the 36-Item Short Form Health Survey (SF36)

Floor and ceiling effects
The floor or the ceiling effect was the same whatever the moment of completion. The floor effect was weak for the two tools except for two dimensions in the SF36: physical role and mental role.

The six dimensions of the NHP were skewed with high scores (either at T1 or T2), indicating the existence of an important ceiling effect, with more than 25% perfect scores at T2. In the SF36, fewer than 25% of scores were equal to 100 in all dimensions (Table 4).

Criterion validity
The correlations between the dimensions of the QOL questionnaires and four disease-specific variables recorded preoperatively are presented in Table 5. Among the sample of variables tested, the strongest correlations were for dyspnea and angina pectoris. Weaker correlations were found between the rest of the preoperative recorded variables and the different dimensions.

Short-time evolution of angina pectoris and dyspnea status
For the SF36 the percentage of well-classified patients at 5 weeks was 69% in the unchanged group and 61% in the improved group. The two most meaningful dimensions were pain (p = 0.0001) and energy (p = 0.0735).

For the NHP, the percentage of well-classified patients at 5 weeks was 70% in the unchanged group and 43% in the improved group. The two most important dimensions were pain (p = 0.0008) and sleep (0.0535).

The weak number of worsened patients (n = 21) precluded analysis in this group.

	Comment

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
A consensus emerging in the literature indicates that mortality and morbidity criteria as outcomes of surgery remain inadequate in many situations [21]. Nevertheless, the results of cardiac operations are still usually assessed by mortality and morbidity rates. While they do give important information about the disease, it is impossible to separate disease from a patient’s personal and social context. The use of QOL measures in clinical practice ensures that treatment focuses on the patient rather than on the disease. QOL instruments help to integrate the patient’s view into clinical practice and into the evaluation of new therapeutic strategies. Nevertheless, it has to be decided how narrowly or comprehensively HRQL should be measured.

In the present study we used two medical criteria strongly correlated to the dimensions of the two questionnaires: dyspnea and angina pectoris. We chose these two criteria because, conversely to comorbid diseases or age, they are supposed to vary between the preoperative and postoperative moments. By using these two criteria we were able to create three medical status groups: improved, unchanged, or worsened. The group of unchanged patients was the largest. This was not surprising as we used a criterion of assessment seldom reported in the literature—the assessment of the short-time QOL to predict the evolution of angina pectoris and dyspnea at 5 weeks.

As there is no specific questionnaire for QOL in cardiac surgery we used two generic instruments, the NHP and the SF36. We built this analysis in order to compare these two instruments to choose one or the other for further studies.

The two instruments used in this study have been compared previously. The authors who examined the acceptability of the NHP and the SF36 [22] reported that, globally, neither the NHP nor the SF36 was judged to be very long or very complex. Both instruments were practical, with more than 96% complete responses, which is in line with our results. Even if the mean time for completion was statistically shorter for the NHP than for the SF36, this difference is not clinically relevant. Some authors favor the SF36 because of its psychometric properties when used in a healthy general population [23, 24]: the SF36 has been described as more responsive than the NHP, thus more useful in the assessment of QOL. Others have criticized the NHP as having a low sensitivity to change, probably due to its use of binary responses (0 or 1) and its propensity toward a ceiling effect [22]. In the current study the SF36 was the more sensitive to change. In the three groups of patients (improved, unchanged, or worsened) the SF36 seemed to perform better than the NHP and reflected the true clinical trend. Concerning the ceiling effect, we obtained the same results as those found in the literature [22, 24]. Obviously, the NHP showed a skewed distribution in the present study, which reflects a considerable ceiling effect. Concerning concurrent validity, we noticed that correlations were rather moderate, probably due to the fact that the questions in the two questionnaires, although they were supposed to measure the same thing, were not asked the same way and were not identically oriented. The internal consistency of the SF36 was excellent in this study, as found in the literature [25], and largely better than that of the NHP. Both scales seemed able to predict the evolution of angina pectoris and dyspnea in the two analyzed groups of patients (improved and unchanged) but the SF36 provided the better results.

Acceptability was in the same range for both questionnaires. The SF36 had generally good psychometric properties and was particularly more sensitive to change than the NHP (so was probably more relevant to a longitudinal study). The SF36 provided better results in the assessment of the evolution of angina pectoris and dyspnea. The SF36 seems more suitable than the NHP for evaluating QOL in cardiac surgery.

	Acknowledgments

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
This study was supported by grants from Program Hospitalier de la Recherche Clinique (Ministère de la Santé Publique), Paris, France.

The authors thank Nancy Richardson-Peuteuil for her editorial assistance.

	References

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 

1. Bowling A. Measuring disease. A review of quality of life measurement scales. Milton Keynes, UK: Open University, 1995.
2. Hunt S.M., McEwen J., McKenna S.P. Measuring health status: a new tool for clinicians and epidemiologists. J R Coll Gen Pract 1985;35:185-188.[Medline]
3. Stewart A.L., Hays R.D., Ware J.E. The MOS short form general health survey. Med Care 1988;26:724-735.[Medline]
4. Chocron S., Etievent J.Ph., Viel J.F., et al. Preoperative quality of life as a predictive factor of 3-year survival after open heart operations. Ann Thorac Surg 2000;69:722-727.[Abstract/Free Full Text]
5. Chocron S., Etievent JPh, Viel J.F., et al. Prospective study of quality of life before and after open heart operations. Ann Thorac Surg 1996;61:153-157.[Abstract/Free Full Text]
6. Fruitman D.S., McDougall C.E., Ross D.B. Cardiac surgery in octogenarians: can elderly patients benefit? Quality of life after cardiac surgery. Ann Thorac Surg 1999;68:2129-2135.[Abstract/Free Full Text]
7. Rumsfeld J.S., McWhinney S., McCarthy M., Jr, et al. Health-related quality of life as a predictor of mortality following coronary artery bypass graft surgery. Participants of the Department of Veterans Affairs cooperative study group on processes, structures, and outcomes of care in cardiac surgery. JAMA 1999;281:1298-1303.[Abstract/Free Full Text]
8. Ware J.E., Brook R.H., Davies A.R., et al. Choosing measures of health status for individuals in general population. Am J Public Health 1981;71:620-625.[Abstract/Free Full Text]
9. Erickson P., Patrick D.L. Guidelines for selecting quality of life assessment: methodological and practical considerations. J Drug Ther Res 1988;13:159-163.
10. Bucquet D., Condon S., Ritchie K. The French version of the Nottingham Health Profile: a comparison of items weights with those of the source version. Soc Sci Med 1990;30:829-835.
11. Hunt S.M., Mc Kenna S.P., Mc Ewen J., et al. A quantitative approach to perceived health status: a validation study. J Epidemiol Comm Health 1980;34:281-286.[Abstract/Free Full Text]
12. McHorney C.A., Ware J.E., Raczek A.E. The MOS 36 Item Short Form Health Survey (SF36). II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:247-263.[Medline]
13. McHorney C.A., Ware J.E., Lu J.F., Sherbourne C.D. The MOS 36 Item Short Form Health Survey (SF36). III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994;32:40-66.[Medline]
14. Leplège A., Mesbah M., Marquis P. Analyse préliminaire des propriétés psychométriques de la version francaise du SF36 dans le cadre du projet IQOLA. Rev Epidémiol Santé Publ 1995;43:371-379.[Medline]
15. Essink-Bot M.L., Krabbe F.M., Van Agt M.H., et al. NHP or SIP: a comparative study in renal insufficiency associated with anemia. Qual Life Res 1996;5:91-100.[Medline]
16. Deyo R.A., Diehr P., Patrick D.L. Reproducibility and responsiveness of health status measures, statistics and strategies for evaluation. Control Clin Trials 1991;12:142s-158s.
17. Katz J., Larson M.G., Philips C.B., et al. Comparative measurement sensitivity of short and longer health status instruments. Med Care 1992;30:917-922.[Medline]
18. Cohen J. Statistical power analysis for the behavioral sciences. New York: Academic Press, 1977.
19. Fayers P.M., Curran D., Machin D. Incomplete quality of life data in randomized trials: missing items. Stat Med 1998;17:679-696.[Medline]
20. SF36 Scoring Rules. Version 1.1. New England Medical Center Hospitals, Inc, 1991.
21. Wood-Dauphinee S. Quality of life assessment: recent trends in surgery. Can J Surg 1996;39:368-372.[Medline]
22. Bouchet C., Guillemin F., Paul-Dauphin A., Briançon S. Selection of quality of life measures for a prevention trial: a psychometric analysis. Control Clin Trials 2000;21:30-43.[Medline]
23. Vanderzee K.I., Sanderman R., Heyink J. A comparison of two multidimensional measures of health status: the Nottingham Health Profile and the Rand 36-Item Health Survey 1.0. Quality Life Res 1996;5:165-174.
24. Brazier J., Jones N., Kind P. Testing the validity of the Euroqol and comparing it with the SF36 health survey questionnaire. Quality Life Res 1993;2:169-180.
25. Goldbeck L., Schmitz T.G. Comparisons of three generic questionnaires measuring quality of life in adolescents and adults with cystic fibrosis: the 36-item short form health survey, the quality of life profile for chronic diseases, and the questions on life satisfaction. Quality Life Res 2001;10:23-36.

This article has been cited by other articles:

				Z. Colak, I. Segotic, S. Uzun, M. Mazar, V. Ivancan, and V. Majeric-Kogler Health related quality of life following cardiac surgery correlation with EuroSCORE Eur. J. Cardiothorac. Surg., January 1, 2008; 33(1): 72 - 76. [Abstract] [Full Text] [PDF]

				B. O. Jensen, P. Hughes, L. S. Rasmussen, P. U. Pedersen, and D. A. Steinbruchel Health-related quality of life following off-pump versus on-pump coronary artery bypass grafting in elderly moderate to high-risk patients: a randomized trial. Eur. J. Cardiothorac. Surg., August 1, 2006; 30(2): 294 - 299. [Abstract] [Full Text] [PDF]

				P. E. Falcoz, S. Chocron, F. Laluc, M. Puyraveau, D. Kaili, M. Mercier, and J. P. Etievent Gender analysis after elective open heart surgery: a two-year comparative study of quality of life. Ann. Thorac. Surg., May 1, 2006; 81(5): 1637 - 1643. [Abstract] [Full Text] [PDF]

				P.-E. Falcoz, S. Chocron, L. Stoica, D. Kaili, M. Puyraveau, M. Mercier, and J.-P. Etievent Open heart surgery: one-year self-assessment of quality of life and functional outcome Ann. Thorac. Surg., November 1, 2003; 76(5): 1598 - 1604. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Sidney Chocron Joseph Philippe Etievent Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Falcoz, P. E. Articles by Etievent, J. P. Search for Related Content PubMed PubMed Citation Articles by Falcoz, P. E. Articles by Etievent, J. P. Related Collections Cardiac - other Related Article