Using a Structured, Computer-Administered Questionnaire for Evaluating Health-Related QOL with Chronic Lower Extremity Wounds
Patients with chronic wounds of the lower extremity (CWLEs) often experience functional disability and emotional distress; incorporating health-related quality of life (HRQoL) measurements in clinical practice may improve understanding of chronic wound patients’ healthcare needs.
A computer-administered instrument that measures HRQoL variables in patients with CWLEs was developed to overcome common limitations to assessing HRQoL in this population. Face validity of the questionnaire variables assessing physical, social, emotional, and functional well-being was obtained and a computer application to display the structured questionnaire on an electronic kiosk with touch-screen interface was developed. All patient responses are stored in the clinic’s electronic health record system (EMR systems). To evaluate use of this system in a wound care clinic, 66 consecutive patients were asked to complete the questionnaire; of those, 64 participated. Internal consistency of the instrument across responses was estimated by the Kuder-Richardson formula 20 as 0.79. None of the patients requested help completing the questionnaire or working with the touch-screen interface. Patients most frequently reported frustration (63%), trouble sleeping (48%), anxiety (42%), and impaired mobility (41%), confirming that CWLEs negatively affect patient quality of life. These findings suggest that additional validation and reliability studies, including research to evaluate the relationship between HRQoL, protocols of care, and wound outcomes, are warranted.
Please address correspondence to: Thomas I. Sherman, 100 Leverington Avenue, PH 15, Philadelphia, PA 19127; email: email@example.com.
The lived experience of chronic wounds evokes powerful emotional issues; more than 25% of patients report symptoms of depression or anxiety.1,2 In a large sociodemographic study3 of 120 patients with venous ulceration, 16% were out of work and 49% were disabled in terms of work tasks. Notwithstanding these circumstances, most clinical assessments of wound healing focus on healing rates and/or morbidity (eg, delayed epithelialization, limb loss, and/or infectious complications). Assessment of functional status and quality of life is fundamental to the care of patients with chronic wounds of the lower extremity (CWLEs) and should not be overlooked.
Health-related quality of life (HRQoL) evaluations examine the psychosocial stresses faced by patients with chronic conditions. HRQoL assessments offer important information throughout the wound care continuum because efforts to improve quality of life and relieve suffering are integral to all wound treatment regimens. The importance of quality of life measurement for patients with CWLEs is gaining recognition.4-6 However, as with other areas of healthcare, progress in this area has not been accompanied by widespread implementation and testing for several reasons. First, HRQoL has been shown to be a complex concept, widely understood to be a multidomain phenomenon with physical, psychological, and social components.7 Second, as described by Price and Harding and observed in the authors’ experiences, many clinicians may be hesitant to regularly employ a generic measurement strategy of HRQoL because such approaches often fail to encompass significant aspects of the condition-specific health state of CWLEs and the clinicians themselves may be unaware of the few measures that have been developed specific to CWLEs, such as the Cardiff Wound Impact Schedule designed by Price and Harding.4 Lastly, the authors often have found in practice that comprehensive and systematic assessment of patients’ psychosocial needs can substantially increase provider time, including carrying out the interview, making appropriate subspecialty referrals, and systematic monitoring as needed.
The utility of patient self-administered health history questionnaires has been studied extensively.8-10 Several reviews analyzing11-13 studies that explored the effectiveness of computer-administered surveys for obtaining patient health information have concluded that dialogue between patient and computer is an effective and well-accepted means of obtaining accurate and comprehensive medical histories. Compared to paper-and-pencil questionnaires and face-to-face interviews, patients have reported a preference for computer-administered surveys, especially when being assessed for psychological histories.14,15 However, a valid, reliable, computer-interpretable instrument for measuring HRQoL in patients with CWLEs does not exist.
The measurement instrument developed by the authors is an electronic, patient self-administered, internally consistent questionnaire that is displayed on a computer kiosk with touch-screen interface and integrated with the clinic’s pre-existing electronic health record (EHR) system for storage. It also has the potential to track patients’ responses each time they receive treatment. The purpose of this prospective descriptive study is to evaluate use of an electronic HRQoL evaluation system in clinical practice.
Literature analysis. To identify the relevant literature, a comprehensive search of English-language articles published between January 1996 and December 2006 was performed of OVID Medline, CINAHL, EMBASE, the Cochrane Wounds Group database within the Cochrane Collaboration Library, the Agency for Healthcare Research and Quality (AHRQ) Clinical Practice Guidelines, and the National Guideline Clearinghouse™ database of evidence-based clinical practice guidelines. The search term combination captured the concepts functional status, quality of life, and wound using a wide range of indexing terms, free text words, and word variants. Bibliographies of key articles also were searched to supplement the literature searches and concise definitions of key terms were adopted (see Table 1).
Development of a preliminary set of items and domains. To identify a set of simple, uncorrelated, closed-ended questions that might measure HRQoL, questions extracted from the medical literature review16-19 were pooled along with a retrospective review of interviews contained in the clinic’s EHR from 724 patient visits to the authors’ ambulatory wound care center.18,19 From the item pool, content domains were ascertained representing four health attributes that assessed HRQoL: physical, social, emotional, and functional well-being. A set of four questions representing each domain for a total of 16 questions was defined (see Table 2). These questions were selected because they were frequently reported in the literature and similar questions are widely incorporated into existing validated health surveys such as the CWLE-specific Cardiff Wound Impact Schedule.4 A yes/no dichotomous response format was chosen for each attribute’s representative question. A positive response to a question is scored as a value of 1 and a negative response as 0; therefore, the scores for each domain can range from 0 to 4. The questions were designed to facilitate a rapid, sensitive, accurate rating of a comprehensive range of relevant domains by naive raters with minimal burden on respondents.
Consensus process and content validity. The first stage of ensuring content validity involved formation of a cross-specialty panel composed of clinical and methodological experts from the fields of nutrition, interventional radiology, anesthesiology, pharmacology, podiatry, vascular surgery, general surgery, plastic surgery, and healthcare policy research to review the instrument and ensure content domains had been sampled adequately (see Table 3). Panel members were informed that the instrument was being designed to measure the physical, social, emotional, and functional aspects of patients’ lived experience with nonhealing wounds. All panel members volunteered their time; no commercial support was provided. The panel also declared in writing their independence from any industry relevant to this topic according to standard reporting requirements.
Consensus was achieved in open forum with foremost consideration given to comments made by participants with greatest expertise on the matter under discussion. When opinions diverged, sources of disagreement were discussed. Throughout the process of collaboration, the group’s aim was to understand the factors that caused disagreement and minimize potential biases. The panel’s overarching goal was to answer four questions: 1) Are the content domains of the instrument salient to the HRQoL status of patients with CWLE? 2) Is the content of the questionnaire fairly estimated from the review of the medical literature? 3) Does the instrument have the potential to meaningfully discriminate between different levels of severity? 4) Would the instrument reasonably appear to be a valid way to monitor the HRQoL for patients with CWLEs if formally tested? Once the panel of experts achieved consensus on these issues, there was extensive discussion as to how the questionnaire could most effectively present these issues to the patient. The final questionnaire was a product of the panel’s expertise and experience in the participants’ respective fields.
Data collection instrument. Next, the paper-based structured questionnaire was converted to an interactive electronic document stored in the clinic’s EHR (see Figure 1).20 To administer the questionnaire, a computer application was developed to “call up” or execute the questionnaire, display it on an electronic kiosk with touch-screen interface (Cintiq, Wacom Technology Corporation, Vancouver, WA), monitor the respondent’s progress, and store each patient’s choices in a relational database. The system provides the ability to de-identify protected health information (PHI) on electronic output while leaving the actual PHI data unmodified in the original record. Before deployment, the application was thoroughly tested by Ryan D. Harlin, information technology consultant for The Wound Clinic (Broomall, PA) for reliability in loading the interface and saving patient data, as well as for overall runtime stability.
The graphic user interface was programmed to be simple, efficient, and immediately comprehensible. When displayed on the touch screen, the digital questionnaire was completed by an easy “see-and-point” process. To ensure survey completeness, the user interface provided immediate feedback when an item was missed (see Figure 2). No time restrictions were imposed and patients were provided a private setting to complete the form.
System engineering respected Nielsen’s21 principles for “usability” and sought to provide a satisfying human-computer interaction. The program that conducted the interview was built in Apple Computer’s Xcode environment using the Cocoa Objective-C application development framework. The customized application was implemented as a three-tiered client/server architecture on the Mac OS X (Apple, Inc., Cupertino, CA) UNIX-based operating system.
Data collection, reliability testing, and scoring. Institutional Review Board approval was obtained from the Delaware County Memorial Hospital (Drexel Hill, PA) IRB for collection of clinical data. The study complied with HIPAA provisions for protecting confidentiality of patient data and the security of EHRs.22 Study participants were seen in a community-based, university-affiliated ambulatory wound care center. All patients who presented for a new-patient visit and care of a chronic wound of the lower extremity were eligible to participate. Sixty-six consecutively eligible patients were approached. The study was explained and 64 patients (97%) agreed to participate and provide informed consent. One patient refused to participate due to concerns about the demands of information technology and the other for unspecified reasons. All participants were asked to describe their current status and instructed to ask the clinic staff for assistance if needed. All sixty-four participants completed the 16-item HRQoL structured questionnaire.
The Kuder-Richardson formula 20 (KR20) was employed to test the internal consistency of the HRQoL instrument. The scoring procedure was straightforward and customary for discriminative HRQoL instruments.23 Because individual items had been formulated to be relatively uncorrelated, scores for each item were equally weighted. The domain scores were summed and then combined into a single overall score.
The mean total score for HRQoL (n = 64) was 12.2 (SD 3.3) (range 1 to 16). Mean scores for each domain were: Physical — 3.0 (SD 1.1), Social — 3.5 (SD 1.0), Emotional — 2.7 (SD 1.2), and Functional — 3.0 (SD 1.2) (see Table 4). When rank-ordered, almost two thirds of patients with CWLEs experienced general frustration, while nearly half reported anxiety and problems sleeping (see Table 5). Impaired mobility affected 41% and more than one third were unfit for work. Wound care services and/or dressing supplies burdened the budgets of approximately 20%. It is interesting to note that although a majority of patients reported frustration, trouble sleeping, anxiety, impaired mobility, and inability to work, only 8% of the participants described themselves as depressed. This finding is closely corroborated by the group’s infrequent reporting of changes in self-confidence (8%) and frequent reporting of compassionate support from family and/or friends (92%).
Validity and internal consistency. Data collection using an electronic kiosk was well accepted by all study participants. None of the subjects requested help of the clinic’s staff despite instruction to do so if needed and no one reported respondent fatigue. No malfunctions in computer hardware or the application created to execute the study occurred. Based on expert interpretation of the scientific evidence and group consensus, the survey questions appear to satisfy the content domains, support for content validity was favorable, and internal consistency was strong, as estimated by the index KR20 = 0.79. Furthermore, the staff at The Wound Clinic did not experience increased administrative burden with this device because the questionnaire was completed by the patients independently in the waiting area and the results stored in the practice’s EHR.
Physiologic measures such as wound-healing trajectories, quality of exudate, and transcutaneous oxygen measurements provide important information to wound care practitioners; patients’ functional capacity and well being are also important. Clearly, measuring HRQoL is an important part of assessing the full burden of nonhealing wounds.
This study is the first report in which an instrument to measure HRQoL was developed and tested for internal consistency using a patient-interactive touch-screen device, integrated with the EHR.26-34 Historically, studies exploring the impact of CWLEs on quality of life have utilized questionnaires administered by trained interviewers25,27-32 or patients themselves using paper forms.24,26 The former method is resource-intensive but ensures compliance and decreases missing items. The latter method relies on the patient’s literacy to understand the questions. The authors believe that none of this study’s subjects had difficulty because no one requested assistance they knew was available. Still, the possibility exists that some patients may have had difficulty understanding the questions and failed to ask for help due to embarrassment or lack of interest. Furthermore, traditional self-administered paper-based questionnaires have been associated with numerous shortcomings, including illegible responses and risk of misplacement.33-37 A computer-administered structured questionnaire minimizes administrator burden, ensures compliance, decreases errors, eliminates missing items, collects uniform data for outcomes research, and provides an opportunity for clinical decision support.11-13
Internal consistency of the data was good. The results also suggest that QoL is affected by a CWLE, especially emotional well-being. This finding is similar to Jones et al,1 who reported 52 out of 190 patients with venous leg ulcers in their study were depressed, as well as to Goodridge et al’s6 cross-sectional study of adult patients with diabetes and foot ulcers in which patients with nonhealing ulcers experience notable anxiety about their wounds.
The importance of disability measurement as part of quality improvement cannot be overstated. It is not possible to improve the quality of clinical wound care without measuring patient responses to therapy. One limitation of this study was that data were collected only once. Hence, sensitivity of the instrument to potential changes or correlations between HRQoL changes and outcomes is unknown. In addition, the cost or resources required to produce a change in HRQoL were not measured and HRQoL results were not compared to those of existing valid and reliable HRQoL instruments.
Instead, the instrument’s creators attempted to derive questions that were content-specific to CWLEs and that would likely prove valid if tested by the literature review, expert panel discussions, and review of the patient interviews in the EHRs. Therefore, no conclusion can be drawn about the overall effectiveness of the scoring system employed or the true differences that exist between patient experiences.
A limitation to the current study was that the instrument was not employed beyond the patient population of one wound care facility, which is located in a suburban setting and the majority of the participants in their sixth decade of life or older. Hence, the utility of this instrument across different geographic populations, residential types, and ages was not assessed.
>Barriers to Implementation
Implementing functional status and quality-of-life assessments can increase demands on provider and patient time. Time pressures and administrative issues already encumber those who provide care to this challenging patient population. Additionally, the inconvenience of measuring quality of life might lessen patient acceptance for repeating the test over time, negatively affecting efforts to measure HRQoL over the full cycle of care.
For quality improvement initiatives to gain real traction in the field of wound care, widespread adoption of health information technology (HIT) is critical. The quality of clinical decisions and confidence in their outcomes will never be better than the quality of information supporting those decisions. However, as described by members of the Institute of Medicine committee and outstanding practitioners, researchers, and policy-makers tasked with implementing healthcare organization and system-focused quality improvement and ensuring means to build quality assessment, many wound care practitioners lack sufficient information system development support for the capture, analysis, and reporting of quality measures.38 Miller et al39 report many impediments to the implementation of EHR systems to improve quality of care, including initial cost and time requirements; furthermore, in their case-report study of 14 solo practices employing EHRs, only two of the practices specifically utilized their EHR to systematically improve chronic care. This is understandable due to the sometimes difficult acquisition and implementation of EHR, as well as sometimes substantial financial burdens and administrative overhead that already cost physicians 40% to 60% of revenues.40 Taken together, widespread HIT adoption may be decades away. In effect, the current situation has been described as “...a period of adoption by the willing.”41 To expect wound care providers to purchase and install information systems may be premature, but the pace at which public and private sector quality initiatives are moving is swift.
Among the obstacles to implementing a measurement of HRQoL is a paucity of rigorous scientific evidence that clearly establishes its value in the clinical care of chronic wounds. Well-designed clinical trials that verify its usefulness and establish an explicit role for standardized HRQoL measurement in this area do not exist. Barring their absence, the development of high-quality decision support systems is beyond reach.
Unfortunately, ideological hurdles must be overcome. The current payment system fails to reward providers who measure results, adopt information systems, and demonstrate excellent approaches to care.42 In fairness, however, it should be noted that steps recently have been made toward ensuring quality of care on this end, such as denied reimbursements for the occurrences of “Never Events” that were devised by the National Quality Forum.43 Yet by and large, industry-wide strategies to bring down costs by reducing provider payments and paring back services effectively subsidize high-volume services and dampen quality improvement efforts. Thus, even if future studies indicated that the formal assessment of the HRQoL data improves treatment, which could be determined by correlation with outcome measures such as healing time, incentives to integrate measures similar to the current endeavor may be negligible. Finally, the prospect of shared knowledge, collaborative processes, and a more egalitarian working environment may be inhibited by providers’reluctance to relinquish authority and power.
The paucity of valid and reliable measurement tools for HRQoL in this patient population calls for methodological standards of instrument development along with computer-assisted cooperative work and sound clinical data analysis. This is an enormous undertaking and requires fundamental shifts in wound care delivery systems. Current programs are in their infancy.
Outcome measurement and reporting likely will play an important role in national initiatives aimed at integrating clinical goals with patients’ values. By providing a basis for developing easily employable HRQoL data collection devices, the study instrument is a step in this direction. More steps will be required for widespread integration into clinical practice, including: 1) determining the external validity of the questionnaire; 2) establishing the sensitivity, specificity, and reliability of the instrument; 3) correlating the outcome measures to changes in HRQoL status for improved patient monitoring and care; 4) determining patient-specific acceptable HRQoL threshold value; and 5) implementing high-quality treatment plans that produce positive changes in nonhealing wound patients’ HRQoL statuses. To truly deliver high standards of care for patients with complex wounds, the authors believe an expert information system and a mature workflow management scheme are imperative. Wound care practitioners need to find their place across the broad rubric of national quality initiatives and define clear strategies to critically evaluate how wound treatments affect patient outcomes.
The instrument described herein demonstrates it is possible to assess the impact of CWLEs on patients’ QoL regarding physical, social, emotional, and functional well-being through an electronic questionnaire and integrate this information into an existing EHR system. This was achieved without increasing the healthcare professionals' administrative burden, as reported by the staff of the clinic in which this instrument was developed. In an age where the benefits of EHR are becoming more readily touted and more commonly employed in practice, it will become increasingly important to find means to seamlessly integrate health information into these records. This instrument provides a basis for future development of devices that can routinely be used to assess HRQoL data from patients with CWLEs that could potentially help improve quality of care, continuously monitor the effectiveness of treatment, and help bring attention to the vast impact that CWLEs have on many facets of patient’s lives.
Dr. Harlin is Clinical Assistant Professor, Department of Surgery, Drexel University College of Medicine, Philadelphia, PA; and Medical Director, The Wound Clinic, Broomall, PA. Mr. Harlin is a rich Internet applications systems developer, Los Angeles, CA. Mr. Sherman and Ms. Shafquat are medical students, Drexel University College of Medicine. Ms. Rozsas is a student at Drexel University, Philadelphia, PA. Dr. Meyers is a Professor and Chair, Department of Surgery, Drexel University College of Medicine.