|Home | About | Journals | Submit | Contact Us | Français|
Chronic obstructive pulmonary disease (COPD) affects over 12 million people and is the fourth leading cause of death in the United States.1 Chronic obstructive pulmonary disease is a slowly progressing lung disease which includes the diagnoses of emphysema and bronchitis. Both emphysema and bronchitis are characterized by patients exhibiting compromised breathing patterns and inefficient movement of air into and out of the lungs. When the airways become obstructed as a result of these disease processes, patients experience excess sputum production, difficulty breathing, coughing, hypoxemia, and hypercapnea.2
While there are many medical treatments used to manage symptoms of COPD, this condition is not curable and slowly progresses despite treatments and medications. Further progression of the disease leads to a progressive loss of function of the pulmonary system and decreased ability to perform previously enjoyed activities. Patients whose disease process leads to decreased functional capacity may experience a sense of loss of self and feel that their quality of life has diminished. Realizing that COPD and other chronic diseases have not only physical, but also emotional and psychological consequences, it is important for health care providers to examine all aspects of patients' health. Limited availability of resources for patient care in today's health care system requires that treatments and interventions used be both cost- and time-efficient, while providing the highest level of care possible for all patients. Various outcome measures are often used to ensure that this level of care is provided. It is no longer sufficient to use only physiologic measures to ensure quality of care as these measures often do not correlate to improved health-related quality of life (HRQL).3–5 Instead, health care providers must also examine how changes in a patient's status and mood may affect their functional ability and how these functional outcomes translate into their HRQL.
Quality of life, as defined by the World Health Organization, is “an individual's perception of their position in life in the context of the cultural and value systems in which they live, and in relation to their goals, expectations, standards and concerns.”6 Health-related quality of life describes a subset of quality of life directly related to an individual's health. Tools that assess HRQL can provide important information about the person's overall health status that otherwise may be overlooked or not taken into consideration. There are currently many outcome tools measuring HRQL, most of which use a questionnaire format and may be completed by an interviewer assisting the patient or by the patient independently. Some tools are targeted towards the effects of a specific disease on HRQL, whereas others are generic and can be used for patients with a variety of diseases. For such a tool to be useful, however, it must be proven to be reliable, valid, responsive, and interpretable.7
The Sickness Impact Profile (SIP) is a measure of health status designed to be broadly applicable across varying types and severities of illness and across an array of demographic and cultural subgroups.8 This article will endeavor to describe the Sickness Impact Profile and assess its usefulness for determining HRQL in patients with COPD
The SIP was developed in 1972 by Bergner8 to provide a measure of perceived health status that was sensitive enough to detect changes or differences in health status which occur over time or between groups. Developers of this generic tool intended to create an outcome measure that could be used for evaluation, program planning, and policy formulation. Field trials of the SIP were conducted in 1973 and 1974 and were designed so that the instrument would be tested on subjects that spanned a range of type and severity of illness. Two hundred forty six subjects consisting of outpatients, inpatients, home health care patients, walk-in clinic patients, and nonpatients completed the SIP.8 In addition, a scoring method was developed and tested. After testing, the number of items was reduced from 312 to 189 and the wording of remaining items was refined. A third field trial in 1976 had 3 aims: determining the final content, format, and scoring of the SIP; providing a broad assessment of the discriminate, convergent, and clinical validity of the SIP; and comparing the reliability and validity of alternative administrative procedures. The final revision of the SIP was based on data from 1976 trial and a cumulative sample of subjects' responses in the 1973, 1974, and 1976 field trials.8
The SIP in its final form contains 136 questions about health-related dysfunction.8 The SIP can be used for any disease condition and does not require specific age or patient characteristics. It can be administered by self-report or by an interviewer. Respondents are asked to mark only those items that are related to their health and that describe the respondents on a given day. Administration of the test takes approximately 20 to 30 minutes and scoring can be completed in approximately 5 to 10 minutes in both self-report and interview forms. No special training is needed to administer or interpret the results. Patients must have a sufficient cognitive level to be able to appropriately answer questions regarding their daily life. Administration to patients with compromised cognitive ability may require use of a proxy. However, the reliability and validity between proxy SIP scores and patient-derived scores were found to be inconclusive.9
The SIP uses a series of questions with 2 domains, physical and psychosocial. Within the domains, there are 12 specific categories that include: ambulation (A), body care/movement (BCM), mobility (M), emotional behavior (EB), social interaction (SI), alertness behavior (AB), communication (C), work (W), sleep/rest (SR), eating (E), home management (HM), and recreation/pastimes (RP). Ambulation (A), body care/movement (BCM), and mobility (M) are included in the physical dimension and all other categories are included in the psychosocial dimension.10
Once the patient has completed the test, a score is determined by summing the scale values of all of the questions in the SIP and dividing that sum by 136. This number is then converted to a percentage. Scores range from 0 to 100. Scores have an inverse relationship with HRQL such that a low score on the SIP is correlated with a high HRQL whereas a high score on the SIP correlates with a low HRQL.8 The final version of the SIP has also been translated into Swedish, German, French, Danish, Norwegian, Japanese, Dutch, and Chicano-Spanish.9
One point of criticism following the development of the SIP was the relatively large number of items contained within the tool. Therefore, a short generic alternative for the SIP was developed by DeBruin et al.11 It contains 68 items divided into 6 categories: somatic autonomy (SA), mobility control (MC), psychological autonomy and communication (PAC), social behavior (SB), emotional stability (ES), and mobility range (MR). The somatic autonomy and mobility control scales together make up the physical dimension of functional status. The psychological dimension includes the psychological autonomy and communication and emotional stability subscales, whereas the social dimension consists of the mobility range and social behavior subscales.12 The total score on the 68-question SIP (SIP68) and its separate categories is calculated by adding the total number of items checked. De Bruin et al11 found that this scoring method produced identical results to the scores obtained on the original SIP version. Scores on the SIP68 range from 0 to 68; and are analogous to the original SIP version where lower scores indicate better HRQL than higher scores.11 The SIP68 has been found to be a reliable measure and is highly correlated with the original SIP (r = 0.94). Researchers believe that it shows promise as a HRQL outcome tool, but conclude that further research is warranted to confirm its usefulness.12 During analysis of the SIP, studies using the tool seldom identified which version of the SIP was used to measure quality of life. Presumably, the original SIP was used, yet the SIP68 may have also been used interchangeably within the studies.
Many studies have examined the usefulness of the SIP in assessing quality of life; however, results between studies have varied somewhat in their findings.3–5, 14–29 While the SIP is a common measure to assess HRQL in a wide variety of patients, its use in patients with COPD has not decisively been determined.
The developers of the SIP examined the test-retest reliability of the SIP by having 6 administrators give the SIP to 119 subjects.13 Subjects included patients receiving rehabilitation and speech services as well as those with chronic conditions receiving treatment in an outpatient setting. Subjects were recruited through convenience sampling and divided into 3 subsamples chosen to cover a range of types of dysfunction and severity of dysfunction. They used both interviewer administration and self-administration of the test. Patients were given the test twice within 24 hours. The period of 24 hours was chosen to minimize the effects of changes in the patient's conditions that would alter their HRQL scores. When comparing subjects' scores, researchers found the test-retest reliability obtained on the 2 administrations of the SIP to be consistent and high (r = 0.88, p < 0.001). In addition to testing the reliability in terms of SIP scores, researchers also examined the test-retest reliability of agreement in terms of items checked.13 Test-retest reliability was performed since the SIP was designed to provide descriptive information on subjects in addition to their scores. An agreement percent coefficient of 0.50 was then computed from these results indicating moderate reliability with regard to the items checked on the tool.13 DeBruin et al12 also found that the overall instrument (psychosocial and physical domains together) as well as the individual categories had high internal consistency as evidenced by Cronbach's a values of 0.94 across all domains.
The SIP has been proven to be reliable among patients with COPD. Janssens14 reported that the SIP is one of the most frequently used instruments measuring HRQL in patients with chronic respiratory disorders. Many of the studies used the SIP as an objective measure of health status and HRQL but did not perform any further analyses of the instrument for populations with chronic respiratory disease. The remaining studies compared the SIP to disease specific outcome tools and lung function parameters. These studies will be discussed later. Based on this lack of information, further research specifically investigating the reliability of the SIP in populations with COPD is warranted.
The SIP was designed to measure the theoretical constructs of health and sickness. The developers of the SIP conceptualized sickness as “changes in behavior associated with the carrying out of one's daily life activities.”9 While disease is objective and illness is subjective, sickness is an inter-subjective construct meaning that sickness can be viewed as a combination of objective findings and subjective feelings. Sickness can be described as the changes in a person's behavior related to his or her health.9 The SIP offers an excellent way to quantify and track changes in a person's objective and subjective feelings. As a result, the SIP can assist clinicians in determining therapeutic treatments that positively affect a client's ability to enjoy the highest quality of life possible.
Because the SIP was developed in part to provide guidance in decision making and policy formation for health care professionals, it is important that validity of the instrument be established. In the absence of a consensually accepted criterion against which alternative measures can be validated, validation research becomes a study of the relationships among measures.15 Since there is no established “gold standard” for assessing HRQL, the criterion validity of the SIP was assessed by comparing patients' self assessment of health status, clinician assessment of health status, and scores on the Activities of Daily Living Index (ADL Index) and National Health Interview Survey (NHIS).15 The SIP was found to have high positive correlation with self-assessments of health status. A correlation of 0.54 was found between self-assessment of sickness and SIP score while a correlation of 0.52 was found for self-assessment of dysfunction and SIP score.15 Statistical significance was achieved despite low correlations between clinician assessments of health status and SIP scores15 (r = 0.49, p < 0.001). The relationship between the score obtained on the ADL Index and SIP overall score was moderately high (0.46) while a high correlation of 0.61 was found between the NHIS and overall SIP scores.15 With regard to the discriminative or descriptive capacity of the SIP, it appears that this tool is capable of describing and delineating groups of illnesses within varying test populations.9,15
The validity of the SIP with regard to pulmonary disease, such as COPD, has been examined in numerous studies with varying results.10,15–17 Engstrom et al10 sought to identify factors that contribute to a diminished health related quality of life (HRQL) in patients with COPD and in doing so, found correlations between SIP scores and a variety of other measures. They used both generic and specific HRQL instruments including the St. George's Respiratory Questionnaire (SGRQ), the SIP, the Hospital Anxiety and Depression Scale, and the Mood Adjective Check List (MACL) to assess HRQL. The 68 participants in the study were separated into groups based on disease severity as measured by forced expiratory volume in one second (FEV1). Pulmonary function tests (PFTs), blood gases, and 6 minute walk test (6MWT) were also assessed in these patients. Results indicated that the SIP ambulation score and the SIP home management scores had the strongest correlations with FEV1 and hence disease severity (r=0.49 and 0.46, respectively). It was also found that the 6 minute walk distance (6MWD) correlated best with SGRQ activity (r= −0.63) and the SIP physical domain (r= −0.67). Of the lung function data obtained from the PFTs, the percent of predicted vital capacity correlated best with the SIP overall scores (r = 0.50) and the SGRQ total scores (r = 0.44). Vital capacity also correlated better with overall SIP scores than with FEV1, although the researchers were unable to determine the reason for this. The correlation between FEV1 and the SIP overall score was r= −0.42 whereas a correlation of r= −0.32 was found for SGRQ and the SIP overall score suggesting fair correlation between these variables. In addition, this study found that there was a strong correlation between the overall SIP and SGRQ total scores.10 This relationship was also observed between most of their respective subscores. On the other hand, SGRQ symptom scores had moderate correlations with the overall SIP and MACL scores with no association found with FEV1. This study reported stronger relationships between lung function and SIP and SGRQ scores on this study than in most other studies. The authors believed that this difference was most likely due to the difference in severity of COPD of the study populations.10 Overall in this study, the SIP was found to have strongest correlations with vital capacity and moderate correlation with FEV1 and SGRQ scores.10
A study by Leyenson et al16 attempted to examine the improvement in HRQL using the SIP in patients with severe COPD after undergoing lung volume reduction surgery (LVRS). They used a case-series analysis approach and examined the effects of LVRS on HRQL and correlated these effects with PFTs, gas exchange, exercise performance, and alterations in medical management. They found that overall SIP scores were lower after LVRS, indicating improvement in health status. Leyenson et al16 also found that the SIP scores correlated with VO2/VE ratio from high to low workloads. The patients having the smallest changes in VO2/VE ratio had the smallest changes in SIP scores after LVRS. Changes in VO2/VE ratio represent decreased energy expenditure during exercise at the same exercise intensity level; therefore, improvements in VO2/VE ratio correspond to improvements in lung function. Researchers also found that improved SIP scores correlated with reduction in residual volume/total lung capacity ratio (r = 0.45, p = 0.09); however, the most significant correlation was found between decreased need for systemic steroids after surgery and improved SIP psychological scores (r = 0.7; p = 0.01). Decreased oxygen requirements post-LVRS also correlated with changes in psychological SIP subscores (r = 0.45, p = 0.09). No correlation was found between SIP scores and routine measurements of lung function, exercise performance, or gas exchange. The authors concluded that changes in HRQL scores on the SIP correlate best with behaviorally based variables such as decreased use of systemic steroids and more efficient breathing patterns during activity that result in improved ability to perform activities of daily living.16 These variables directly affect a patient's well-being and therefore, improve their HRQL. After LVRS, 35% of patients in the study showed a reduction in overall SIP score, close to 50% showed an improvement in the physical score, and nearly 45% had an improvement in psychological score.16 Good validity of the SIP for use with COPD is suggested based on the correlation with improved VO2/VE ratios, improved residual volume/total lung capacity ratio, and decreased use of systemic steroids and supplemental oxygen after LVRS.
A study by Cordova et al17 examined long-term stability of improvements in exercise capacity and HRQL as measured by the SIP post-LVRS in patients with severe COPD, as determined by FEV1 and hyperinflation (total lung capacity). These patients were assessed using PFTs, symptom-limited cardiopulmonary exercise tests, 6MWD, and the SIP pre-LVRS and again at 3, 6, 12, and 18 months post-LVRS. Researchers found that mean SIP scores were significantly lower 3 months after LVRS (score of 7 vs. 18). As expected, mean subscores of the SIP also decreased post-LVRS, with physical scores decreasing from 13 to 4 (p < 0 .008) and psychosocial scores decreasing from 11 to 4 (p < 0.02). In this study, subjects also experienced improvements in lung function, exercise performance (VO2 max), VEmax, breathing pattern efficiency, and respiratory muscle function specifically diaphragm strength.17 The SIP can be inferred to be valid as a measure of HRQL since SIP changes are associated with improvements in lung function, exercise performance, VE max, breathing pattern efficiency, and respiratory muscle function that correlated with improved SIP scores.
Validity of the SIP has been greatly researched to assess its ability to be used as a clinical tool. In summary, the SIP has been proven to be a valid tool for assessing HRQL in the general population by comparing it to other tools related to HRQL. It has high correlation with self assessment of health and the NHIS. The SIP also has a moderately high correlation with the ADL Index.15 When examining the correlation of SIP scores to well researched PFTs to determine its use for patients with COPD, the SIP ambulation and home management scores are closely related to FEV1; whereas, the overall SIP score better correlates with the measure of vital capacity.10 Dividing the SIP into its 2 domains also helps to increase the validity of the SIP outcome tool. The 6MWD shows a strong relationship with the physical domain of the SIP.10 Meanwhile, the psychosocial SIP score correlates highly with decreased oxygen requirements analyzed post LVRS as well as with decreased use of systemic steroids.16 Overall, when the patients are able to breathe easier and use their lungs to optimal capacity, the scores on the SIP improved. With no “gold standard” currently established for assessing HRQL, researchers continue to validate HRQL tools by comparing them to each other and to measures of factors that likely affect HRQL. More research is needed to develop an accepted standard for assessing HRQL.
The standard error of measurement (SEM) is a form of measurement that evaluates response stability by estimating the standard error in a set of repeated scores.18 While the MDC of the SIP has not been calculated specifically for people with COPD, the MDC was determined based on data collected from a study by MacKenzie et al.19 This study used the SIP to analyze change in HRQL in 22 acute medical patients and 21 elective surgical patients with a mean age of 54.5 ± 16.3 years old. These subjects were initially assessed using the SIP prior to beginning medical treatment. They were given 4 follow up interviews with the SIP postdischarge. The error of measurement was determined to be within a confidence interval of 95% for the subjects interviewed. The scores that were obtained for both the psychosocial and physical domains had to lie outside 2 standard deviations to be considered improved or deteriorated according to the SIP.19
For the change in score to be considered meaningful, the results indicated that the changes in scores used to assess physical functioning needed to be greater than −12.5 points to indicate improvement while a change of more than 10.3 points was needed to indicate deterioration.19 For the psychosocial domain, improvement would have to be greater than −16 points and deterioration more than 11.4 points for the change to be meaningful.19 Differences in SIP scores would have to be outside of these ranges for a variation in the subject's HRQL to be detected.
Based on this information, if a score on the SIP lies within the SEM for the specific domain (physical or psychosocial), then conclusions can be drawn that the score may not represent a true change because it falls within the measurement error. There is a lack of conclusive research determining the minimal detectable change of the SIP for patients with COPD and more research is needed to establish this value in this population.
The SIP is a tool that quantifies impairment of health over a wide spectrum of diseases, therefore; relevant content to COPD may be low. The distribution of SIP scores in those who have a diagnosis of COPD (lower with mild-moderate COPD, higher with severe lung disease) indicates that these patients have mild to moderate impairment in HRQL and suggests that the SIP is more responsive to decline in health status compared to improvements in health status.20
“Responsiveness” or “sensitivity to change” refers to the likelihood of detecting a clinically important change or treatment effect.9 Mackenzie et al19 found that the SIP was responsive to large changes for the worse, in medical and surgical patients, with initially mild dysfunctions in the psychosocial domain. Jones21 found that the SIP provided reliable estimates of health in airway disease, but in clinical trials the SIP failed to detect improvements in health despite improvements in other clinically relevant measures. In patients with severe COPD, the use of transtracheal oxygen was associated with a marked reduction in hospital days and improvement in walking distance; however, there was no improvement in SIP scores.22 In a study administered by the Nedocromil Sodium Quality of Life Study Group,23 it was found that the SIP failed to detect an improvement in health, but found a clinically significant improvement in health using the SGRQ.
The aforementioned studies support the evidence that the SIP has been found to be responsive to changes in patients with severe respiratory disease; however, it is not discriminative of mild or moderate cases of COPD. The SIP, however, can be an effective tool in tracking a patient's progression of COPD. If a health care professional notices an increase in a patient's SIP score, it could be an indication that the disease may be progressing and another course of treatment may be warranted.20–23
When examining the usage of SIP with any disease process, previous studies have not been able to clearly establish minimally clinically important differences (MCID). Currently, there is no gold standard or accepted way to assess MCID or to compare the responsiveness of different health-related quality of life instruments.24 It has been suggested that a change of 5 points on the SIP is considered of MCID.25 A change in 5 points is needed to detect MDC in the physical domain as compared to needing to observe a change of 8 to 11 points in the psychological domain. It is imperative when dealing with patients diagnosed with COPD to identify external criteria to help interpret the magnitude of change when using the SIP.26 This can be accomplished by conversing with the patient to identify the smallest change that is important to them, their family, and clinician. The most assured way to identify the MCID is to repeatedly use the SIP with COPD patients.24
The instrument appears to be applicable in a wide variety of patient diagnoses. Most studies, however, focus on chronic conditions.9 It is believed that the SIP accurately describes and defines different populations, and has the ability to differentiate between varying levels of functional status within a diagnosis group.9 The instrument seems better fit for cross-sectional designs than for longitudinal designs.
High levels of dysfunction have been shown in COPD patients with an FEV1 < 50% of normal predicted values.27 Sickness Impact Profile, therefore can be a useful tool when dealing with those who have a diagnosis of severe COPD. The SIP allows clinicians to identify problems that are unique to each patient, tailor a program specific to the patient's needs, and constantly evaluate the effectiveness of treatment in those whose values are less than 50% of FEV1.27
The studies by Leyenson et al16 and Cordova et al17 found that there was a relationship between decreased SIP scores and implementation of a treatment program consisting of alterations in medical management, PFT, and limited cardiopulmonary performance in those who had a lung volume reduction surgery (LVRS). While LVRS is a last resort treatment option in COPD, this information can be valuable when devising a treatment protocol for someone whose disease process is not as severe. By identifying programs unique to the patient, the clinician is better able to work towards goals based on the patient's view of his or her quality of life as compared to making a decision based primarily on diagnosis.28 When tailoring a program specific to the patient's needs one must also consider the psychological component of the SIP. This can be accomplished by incorporating standard psychological interventions into rehabilitation and patient education programs. More patients may be inclined to participate due to this form of therapy rather than participating in a traditional psychotherapy protocol. This type of treatment program would consist of educational lectures and structured group exercises for those who have a diagnosis of COPD. The program would also incorporate their families, by addressing the psychosocial affects of COPD and how they can cope together with the patient.29 Health related quality of life as measured by the SIP score can assist clinicians and patients in determining the effectiveness of treatment.
The purpose of this paper was to evaluate the usefulness of the SIP in assessing HRQL in those who suffer from COPD. After reviewing the literature, it has been consistently shown that the SIP is not the best predictor of HRQL in patients with mild to moderate COPD.10,16,20–23 Therefore, it is of clinical importance to use a disease specific HRQL measurement tool to obtain a true assessment of the level of dysfunction in those who have a diagnosis of mild to moderate COPD. On the contrary, SIP has been proven to be an acceptable tool in determining the HRQL in those who have a diagnosis of severe COPD and have a known FEV1 < 50%.27 The SIP is a useful tool for physical therapists, because it can quantify the effectiveness of rehabilitation treatments, identify decline in condition before symptoms are present, and assist in the development of treatment protocols for targeted patient populations.