The PC-SAD is a Patient Administered Depression Screening Instrument. As many as 50% of patients with major depression seen in primary care clinics are not diagnosed. To facilitate efficient identification of primary care patients with depression, we developed a new patient- administered depression screening instrument (PC-SAD) that produces a DSM-IV diagnosis.
The PC-SAD consists of 37 items: a three item pre-screen, a 26-item MDD section, and an eight-item dysthymia section. The 26-item MDD section includes five questions from the Short Form-36 (SF-36). Questions are laid out in grid formats. The three-item pre-screener consists of two depression questions that are closely related to the WHO depression screener, and one dysthymia question that we introduced. The two pre-screener questions were posed in terms of 1-week recall period. The pre-screen questions become part of the scorning algorithm if a patient's reply is yes. Otherwise, this reduces respondent burden by terminating the screener when all are negative.
We compared its performance to other screeners that yielded DSM-IV diagnoses. To assess validity, the diagnostic accuracy of the PC-SAD was compared with the Inventory to Diagnose Depression (ODD) and the PRIME-MD-PHQ (PHQ) in a convenience sample (n=312) of health plan members, primary care outpatients, and psychiatric patients with diagnoses. The screeners were compared with each other and with psychiatric diagnoses to assess their relative performance.
Disagreement among the screeners was formally tested using a triangulation approach that incorporates a statistical likelihood model. The performance of the PC-SAD and the IDD were comparable. The PHQ was less sensitive than either of those. The PC-SAD respondent burden strikes a balance between the very short PHQ, and the longer IDD, and has the lowest (easiest) Flesch-Kincaid reading level. Investigators, clinicians, and health plans that want a DSM-IV-based depression screener can choose from any of these instruments, with known tradeoffs in sensitivity, respondent burden, and readability.
To read more about the validity, reliability, sensitivity, and specificity, consult the following publication:
- Rogers WH, Wilson IB, Bungay KM, Cynn DJ, Adler DA. Assessing the performance of a new depression screener for primary care (PC-SAD). Journal of Clinical Epidemiology 55 (2002) 164-175.
For questions or to obtain the instrument, call 1-617-636-8078 or email Doris Hernandez
Clinical researchers, the pharmaceutical industry, employers, managed care organizations, and public health professionals are all seeking accurate information concerning the work and productivity impact of employee health problems.
The pharmaceutical industry requires sensitive and specific work disability and productivity indicators for use in its clinical trials and to provide useful information to employers and other health care system stakeholders. Employers are requesting data to assess the impact of changing employee demographics on health and productivity, to evaluate the need for health and work productivity improvement strategies and, once implemented, to evaluate their impact. Managed care organizations are being asked to demonstrate "value" to customers, including purchasers, who are interested in improving employee function and limiting the indirect costs of illness. Finally, public health officials have widened their surveillance and prevention efforts to include disability due to chronic disease. The WLQ can contribute information to all of these initiatives.
The WLQ is an easy to use questionnaire, available in the original 25 item version and an 8-item version. Generally, the 25-item version is used in research and evaluation. The 8-tem version is well suited for inclusion in health assessment tools and other applications that benefit from a short form. The WLQ measures the degree to which employed individuals are experiencing limitations on-the-job due to their health problems, and health-related productivity loss (Presenteeism). The WLQ items ask respondents to rate their level of difficulty or ability to perform specific job demands.
The job demands, which are contained in the WLQ's items, have four defining features:
- They occur among a variety of jobs
- Many different physical and emotional health problems may interfere with their performance
- They are considered important to the job from the worker's perspective
- Problems performing them are frequently related to productivity.
The WLQ's 25 items are aggregated into four scales. The Time Management scale contains five items that address difficulty handling time and scheduling demands. The six-item Physical Demands scale covers a person's ability to perform job tasks that involve bodily strength, movement, endurance, coordination and flexibility.
The Mental-Interpersonal Demands Scale has nine items addressing cognitive job tasks, and on-the-job social interactions. The fourth scale is the Output Demands scale and it contains five items concerning diminished work quantity and quality.
Scale score range from 0 (limited none of the time) to 100 (limited all of the time) and represent the reported amount of time in the prior two weeks respondents were limited on-the-job. Additionally, using an algorithm, WLQ scale scores can be converted into an estimate of productivity loss.
The WLQ development process began in 1994 with a grant from Glaxo-Wellcome, Inc. WLQ research has also been supported by Pharmacia, Inc., Pfizer, Inc., and the National Institute of Mental Health. Prior to the WLQ's development, there had been a very limited amount of detailed information available on health-related work productivity loss. Much of the information was gleaned from global indicators, such as the activity limitation and disability day items appearing within the US National Health Interview Survey, and the role disability scales of health assessment questionnaires (2).
The WLQ itself evolved from a qualitative and quantitative research process. Early in that process, we convened multiple focus groups, consisting of employed patients with chronic disease. These interactions with working patients helped us to better understand how the work activities, associated with various jobs, were influenced by different conditions and their treatments. For example, we found that several physical and mental conditions made it difficult for individuals to perform their job tasks effectively throughout the workday according to an established or expected work schedule.
Another important finding was that work productivity was a sensitive topic for many of the chronically ill workers we interviewed, and we learned how, in a non-threatening manner, to ask about work productivity. We also found that the act of recalling information about work productivity, and the effects of health problems on productivity, constituted a relatively difficult response task. These general findings helped to shape our measurement approach.
A period of cognitive testing followed, in which items and item groupings were evaluated for content validity (relevance to work and to illness), clarity, and respondent burden. Finally, a series of psychometric tests, conducted on the resulting questionnaire forms, led to the current 25-item version. Within patient and employee populations, this version of the WLQ has demonstrated excellent scaling properties, as well as construct and criterion validity.
Scale alpha's exceed the recommended level of .70 in both patient and employee populations. Construct validity tests have shown that WLQ scale scores vary with SF-36 measures of physical and mental health, type of chronic condition, and severity within condition groups, such as depression and osteoarthritis.
Criterion validity tests have been performed in several settings. For example, within a sample of private short-term disability claimants with back pain, baseline WLQ scores obtained within four weeks of the claim predicted the duration of the disability until return to work. In a study of patients with rheumatoid arthritis and in a second with a fibromyalgia sample, WLQ scores predicted patient income level. The WLQ has also been shown to correlate with adverse events in the workplace such as employee injuries.
In a work-site study involving repeat measures of approximately 900 employees, WLQ scores were significantly related to objectively-measured employee-level work productivity. Using results generated within this study, we developed and validated an approach to scoring the WLQ, which enables the user to translate scale scores into a single estimate of productivity loss. The WLQ Productivity Index Score indicates the percentage difference in output from a healthy (not limited) benchmark population.
National Benchmarking Database
In 2003, we completed our first national WLQ benchmarking survey. This survey involved a representative sample of adults in the United States and will generate high quality benchmarks for the population and important subgroups (defined by health condition, occupation and other demographic features). Data from this survey will become available to WLQ users in 2005.
The WLQ is available in:
- Original form (25-item)
- Short form (8-item)
- Web, Phone and Mail Versions
- English (U.S.) and over 30 official language translations
- With a Work Absence module
- With other modules measuring Work, Health, Medical care and Demographic characteristics
WLQ licensing and fee information is available upon request.
For information about the WLQ and consultation services, please contact us by email or call (617) 636-8636.
Work Limitations Questionnaire, © 1998, The Health Institute, Tufts Medical Center f/k/a New England Medical Center Hospitals, Inc.; Debra Lerner, Ph.D.; Benjamin Amick III, Ph.D.; and GlaxoWellcome, Inc. All Rights Reserved.