NIH PROMIS Smoking
NIH PROMIS for Smoking is an electronic standard assessment patient reported subjective outcomes. The electronic delivery is often beneficial compared to paper assessments as they reduce data entry errors and increase enrollment. Qolty PROMIS Smoking is the ePRO system, and can be configured as a PRO tool that provides a short and reliable assessment for smoking with only the most informative items for an individual patient from an item bank based on the prior answers given by the patient.
PROMIS’s self-assessment toolkit for current adult smokers measures six constructs: nicotine dependence, coping expectancies, emotional and sensory expectancies, health expectancies, psychosocial expectancies, and social motivations. Assessing smoking-related constructs is difficult to compute due to the various aspects involved in its measurement, but through the use of computer analyzed tests, short forms, and item banks, an efficient, flexible, and versatile assessment toolkit for sustained use in cigarette smoking research was created. The ultimate hope is to improve the understanding of smoking and cessation.The item banks and short forms assess cravings or withdrawals; the tendency to use smoking to cope; how smoking improves cognitive abilities; and the social benefit of smoking.
The questionnaire was developed after assessing over 5000 daily and nondaily smokers about their opinions regarding smoking. In order to be eligible for the study, participants had to be at least 18 years or older, had been smoking cigarettes for at least one year, had smoked a cigarette in the last thirty days, and had no plans of quitting within the next six months.
PROMIS’s smoking assessment toolkit consists of six constructs essential to comprehend smoking research. These 6 constructs are further divided into three sub-categories: all smokers, daily smokers, and nondaily smokers. Nicotine Dependence (20; 27; 27 items), Coping Expectancies of Smoking (11; 15; 18 items), Emotional and Sensory Expectancies of Smoking (15; 16; 17 items), Health Expectancies of Smoking (12; 19; 18 items), Psychosocial Expectancies of smoking (14; 20; 15 items), and Social Motivations for Smoking (7; 12; 12 items).
All of these items are self-filled and scored with two different types of labels: one relating to frequency (never, rarely, etc.), and the other relating to the intensity (not at all, a little bit, etc.). Now, Qolty makes assessment more accessible, as patients can complete the surveys simply by using their SmartphonesThe PROMIS Smoking Initiative evaluates the relationships between six major constructs and how it relates to or affects smokers. Nicotine Dependence assesses cravings or withdrawal that occurs after brief cessation; i.e. “When I run out of cigarettes, I find it almost unbearable.” Coping Expectancies assesses smokers’ tendency to use smoking to cope; i.e. “I rely on smoking to deal with stress.” Emotional and Sensory Expectancies assesses perceptions of improved cognitive abilities, pleasurable sensations, etc.; i.e. “I feel better after smoking a cigarette.” Health Expectancies assesses smokers’ perceptions of current and long-term health effects; i.e. “Smoking is taking years off my life.” Psychosocial Expectancies assesses smokers’ feelings of social disapproval, normative values, etc., towards smoking; i.e. People think less of me when they see me smoking”. Social Motivations assesses the social benefit of smoking and social cues that induce craving; i.e. “Smoking makes me feel better in social situations.”
The purpose of the questionnaire is to help researchers understand a person’s desire to smoke and the reason(s) why he or she does it. The ultimate goal, as is the same for nearly all smoking-based research, is to help determine a way to end the addiction. Smoking is still one of the leading causes of death and eradicating the addiction is the desired intention behind this assessment toolkit.Similar assessments for smoking are available to people, but none are quite as substantial as PROMIS’s Smoking Initiative and merely determine whether or not one has an uncontrollable problem with smoking and cessation. For example, physicians have used the CAGE questionnaire, the “four Cs” test, and the Fagerström Test to determine nicotine dependence for years (Rustin, 2000). These tests ask simple questions (do you smoke within half an hour of waking up; do you ever smoke more than you intend; how many cigarettes do you smoke each day; etc.) but do not delve deeper into the reasoning why. As mentioned, these assessments, for the most part, are used solely to determine whether or not one has an addiction and all fall short in comparison to the extensive assessment toolkit designed by PROMIS.The Smoking Assessment Toolkit has been successfully translated into Spanish, which includes all of the item banks and short forms for daily smokers. Evaluations of translated version have yet to be completed and will be available in the future.A sample data depicts the average of adults who currently smoke or have smoked in the past. A little more than half of the adults claim to have never smoked in their lives, but nearly half of them have at least smoked one cigarette in their life.
Marginal reliability estimates were compared across the various short forms. The study compared the test information curves (which are an indication of score precision) of candidate forms. The assessment corresponded to a score reliability of 0.80 for respondents up to three SDs above and below the daily smoker population mean (Hansen, 2013).
Based on available information from the fitted item bi-factor model, 8 of the 23 items were excluded from the item banks due to their low computed relationships to the constructs (Hansen, 2013).
The majority of items in a bank for a given domain are identical (with only a few items specific to smoker distinction), and this high correspondence in item content allowed for the creation of short forms that could be administered to all smokers, regardless of daily/nondaily status (Edelen, 2014).
Future studies include administering smoking short forms to daily and nondaily smokers in a community setting. This study will also include a test-retest substudy in which the subjects will be administered two tests within about a week’s time. This is particularly important because the current estimates of test information and marginal reliability depend solely on internal consistency (Hansen, 2013).
Because items in CAT tests are tailored to the particular respondent, these adaptive tests often achieve high levels of score precision (Hansen, 2013).
The item banks for daily smokers contain up to 27 items each and produce scores with estimates above 0.90 for a wide range of each cigarette smoking domain continuum (Edelen, 2014).
Short form and CAT administrations exceed this by achieving reliability scores greater than 0.85 despite having fewer items (Edelen, 2014).Strengths
The psychometric evidence gathered from the item banks, short forms, and CATs of daily smokers provided researchers and clinicians with an array of highly reliable approaches to the six key smoking constructs (Edelen, n.d). The answers provided by smokers to the item banks produced a score with significantly high-reliability estimates. In addition to this, the answers provided for the short forms and CAT administrations gave extra support to the findings of the item banks, with similar results found during nondaily assessments.
First, the performance of both short forms and computerized adaptive tests were evaluated using simulated respondents (Hansen, 2013). Thus, it is unclear how closely related the results in practice would resemble those presented during the simulation. Additionally, the balance of content was considered during the formation of the short forms but was largely ignored during the CAT simulations. This unstructured, unbalanced assessment simulation could have negatively affected the results obtained through the study. The biggest limitation, however, was the complexity of developing separate item banks in each domain for daily or nondaily smokers. Testing the six constructs only could exclude other key factors behind smoking.
- This NIH initiative, made available to the public by PROMIS, was developed with the ultimate goal of understanding smoking and cessation. An assessment toolkit was established after some serious research was done with daily and nondaily smokers. Through the extensive data collected from these two groups, a comprehensive, one-dimensional assessment platform for smokers was created. This assessment focuses primarily on six constructs that researchers believe to be of central importance.
- The six constructs being focused on are nicotine dependence, coping expectancies, emotional and sensory expectancies, health expectancies, psychosocial expectancies, and social motivations
- Participants of the study must be at least 18 years old, have been smoking for a year, have smoked a cigarette in the last thirty days, and have no plans of quitting in the next six months.
- Smallest item bank includes 7 items, and the largest item bank includes 27.
- Assessment test can be used by any smoker, regardless of daily/non-daily status.
Edelen, M. O., Tucker, J. S., Shadel, W. G., Stuck, B. D., & Cai, L. (2012) Toward a more systematic assessment of smoking: developing a smoking module for PROMIS®. Addictive Behaviors, 37,1278-1284. doi: 10.1016/j. addbeh.2012.06.016
Edelen, M. O. (2014, September). The PROMIS smoking assessment toolkit–background and introduction to supplement. Nicotine Tob Res;16 Suppl 3: S170-4. doi: 10.1093/ntr/ntu086.
Edelen, M. (n.d.). Item Banks and Short Forms. Retrieved July 19, 2017, from http://www.rand.org/health/projects/promis-smoking-initiative/item-banks-short-forms.html#files
Rustin, T. A. (2000, August 01). Assessing Nicotine Dependence. Retrieved July 20, 2017, from http://www.aafp.org/afp/2000/0801/p579.htm
Hansen, M., Cai, L., Stucky, B. D., Tucker, J. S., Shadel, W. G., & Edelen, M. O. (2014, September). Methodology for developing and evaluating the PROMIS smoking item banks. Nicotine Tob Res;16 Suppl 3: S175-89. doi: 10.1093/ntr/ntt123
Stocking, M. L., & Swanson, L. (1993). A method for severely constrained item selection in adaptive testing. Applied Psychology Measurement, 17, 277-292. doi: 10.1177/014662169301700308
van der Linden, W. J., & Reese, L. M. (1998). A model for optimal constrained adaptive testing. Applied Psychological Measurement, 22, 259-270. doi:10.1177/01466216980223006