A validated tool for gaining insight into clinicians' preventive medicine behaviors and beliefs: The preventive medicine attitudes and activities questionnaire (PMAAQ)
Introduction
Behavioral risk factors such as smoking, poor diet, physical inactivity, alcohol consumption, and risky sexual activity were the suspected underlying cause of mortality in 48% of all U.S. deaths in 2000 (Mokdad et al., 2004). Clinicians play an important role in mitigating preventable health problems. Their relationship with patients enables them to assess risk factors, effect behavioral change, recommend screening, and prescribe appropriate chemoprophylaxis (Grundy et al., 1997, US Preventive Services Task Force, 1996, Whitlock et al., 2002). Yet reports suggest that clinicians provide preventive services to their patients at less than optimal rates (Finney Rutten et al., 2004, Ma et al., 2004, Natarajan and Nietert, 2003, Nelson et al., 2002). At ambulatory visits in 2000, patients with hyperlipidemia, hypertension, obesity, or diabetes received sub-optimal levels of dietary and physical activity counseling (provided at only 39% and 26% of visits, respectively) (Ma et al., 2004). During the 1990's, screening rates for cervical cancer, colorectal cancer, and cholesterol were unchanged or declined in the majority of states (Nelson et al., 2002). Patients have cited lack of awareness of needed tests and lack of recommendation from their doctor as major barriers to getting Pap smears, mammography, and colorectal cancer screening (Finney Rutten et al., 2004).
If improvements in the delivery of clinical preventive services are to be realized, more must be learned about factors affecting clinicians' behaviors and attitudes. Are clinicians comfortable counseling patients about alcohol consumption or risky sexual behaviors? How important do they think it is to assess the dietary and exercise habits of their patients, overweight and non-overweight? Do clinicians believe their efforts to promote smoking cessation are effective? What specific barriers to preventive service delivery do clinicians encounter, and which are perceived as most important? Few validated surveys have been developed to assess factors such as these (Ely et al., 1998, Gemson et al., 1995) despite their potential to inform targeted interventions for healthcare providers, assess intervention efficacy, and prompt changes to clinician education programs or care delivery systems.
The aim of this study was to validate the Preventive Medicine Attitudes and Activities Questionnaire (PMAAQ). The PMAAQ assesses clinicians' self-reported prevention behaviors (e.g., risk assessment, behavior change promotion, lifestyle counseling); perceptions about the effectiveness and importance of these activities; comfort with addressing sensitive topics with patients such as drug use or sexual behavior; and perceived barriers to the delivery of clinical preventive services. In an earlier publication, we reported on the validity and reliability of three cardiovascular disease (CVD) subscales of the PMAAQ (Murphy et al., 2000). Here, we report validity, reliability, and stability results for the complete instrument. To improve generalizability, we expanded our validation sample – originally limited to University of Minnesota family medicine residents – to include respondents from five other geographically dispersed primary care residency programs.
Section snippets
PMAAQ development
The PMAAQ has its origins in the Clinician Preventive Care Form, a tool for evaluating the United States Office of Disease Prevention and Health Promotion's office system, Put Prevention Into Practice (McGinnis and Griffith, 1996, Yeazel et al., 2002). We grounded the PMAAQ in the Care Form to allow comparisons with results derived from earlier evaluations. Although the theoretical underpinnings of the Care Form are unknown to us, key constructs of that tool and the PMAAQ are consistent with
Results
Descriptive statistics for the PMAAQ scales and subscales are in Table 4. All eight major scales are internally consistent, with high α coefficients (0.74 to 0.98).
Correlation matrix scores are in Table 5. The scales show good divergent validity. Of the 28 correlations, seven range between 0.31 and 0.54. The highest of these, which are all correlated with the Overall Prevention Behavior scale (0.41 to 0.54), are the two other behavior scales (Smoking Cessation, Hypertension Management) and the
Discussion
Results demonstrate that the PMAAQ has good validity and reliability. Content validity was established by experts in preventive medicine and community health. Scales and subscales have sufficiently high Cronbach's α, a measure of internal consistency reliability. Scales were found to have relatively low intercorrelations, indicative of divergent validity (i.e., that the scales are independent of one another and truly measure different constructs). Moreover, several of the scales/subscales were
Conclusion
The complete PMAAQ, and its individual scales/subscales, have validity and reliability for assessing primary care residents' self-reported behaviors, attitudes, and perceived barriers to the delivery of clinical preventive services. Potential applications include model building, needs assessment, intervention evaluation, or quality assurance.
Acknowledgments
The authors thank Anne Marie Weber-Main and Heather Haley for their critical review and editing of manuscript drafts; David McCaffery and Kim Murphy for help with data analysis; Elizabeth Greene for manuscript preparation; and the residency programs at Dartmouth, Arkansas, Texas Southwestern, and Tulane, as well as the Texas Department of Health, for sharing their survey data. This study was supported financially by a Health Resources and Services Administration Preventive Medicine Graduate
References (24)
- et al.
Performance change in an organizational setting: a conceptual model
J. Contin. Educ. Health Prof.
(1997) Social Learning Theory
(1977)- et al.
How individuals, environments, and health behavior interact
Coefficient alpha and the internal structure of tests
Psychometrika
(1951)- et al.
The effect of physician characteristics on compliance with adult preventive care guidelines
Fam. Med.
(1998) - et al.
Examination of population-wide trends in barriers to cancer screening from a diffusion of innovation perspective (1987–2000)
Prev. Med.
(2004) - et al.
Putting prevention into practice. Impact of a multifaceted physician education program on preventive services in the inner city
Arch. Int. Med.
(1995) - et al.
Measuring physician behavior
Med. Care
(1986) - et al.
Changing provider behavior: an overview of systematic reviews of interventions
Med. Care
(2001) - et al.
What drives change? Barriers to and incentives for achieving evidence-based practice
Med. J. Aust.
(2004)
Guide to primary prevention of cardiovascular diseases. A statement for healthcare professionals from the task force on risk reduction
Circulation
Risk reduction: attitudes and behavior of family practice residents
Fam. Med.
Cited by (29)
Adaptation of the Drug and Drug Problems Perception Questionnaire to assess healthcare provider attitudes toward adolescent substance use
2019, Preventive Medicine ReportsCitation Excerpt :Therefore, one use of the ASUPPQ may be to assess provider attitudes prior to training to determine provider readiness for learning about screening tools and practices and/or if a differentiated training curriculum is needed to target attitudes and self-efficacy first for some providers. Indeed, measures used before training initiatives to assess needs and baseline readiness to implement an innovation often include provider beliefs or attitudes about the topic or innovation among other indicators such as provider knowledge, practices, technical capability and organizational factors (Watson et al., 2003; Wilson et al., 2004; Yeazel et al., 2006). The ASUPPQ could also be used to identify opportunities to pair practitioners based on their perspectives about adolescent substance use to optimize peer learning and support throughout the training and implementation process.
Web-based training for primary care providers on screening, brief intervention, and referral to treatment (SBIRT) for alcohol, tobacco, and other drugs
2014, Journal of Substance Abuse TreatmentCitation Excerpt :To assess screening practices, participants were asked, "During the past 90 days, with an adult patient during a periodic health maintenance visit or routine check-up, how often did you do the following…" To assess brief interventions and referral for alcohol and other drug use, participants were asked, "During the past 90 days, when you saw a patient who was drinking heavily, abusing prescription drugs or using illicit substances, how often did you…" To assess brief interventions and referral for tobacco use, participants were asked, "During the past 90 days, when you saw a patient who was a smoker or tobacco user, how often did you…" To assess behavioral intentions, the questions adapted from the Preventive Medicine Attitudes and Activities Questionnaire (Murphy et al., 2000; Yeazel et al., 2006) to assess clinical practices were adapted further. Stems were modified to say, "In the upcoming 90 days…how often do you intend to…" Otherwise, the items were the same.
Are provider self-efficacy and attitudes related to cardiovascular prevention associated with better treatment outcomes?
2007, Translational ResearchCitation Excerpt :The resulting survey consisted of (1) 6 items on the importance of getting to recommended treatment goals (eg, “Lowering a patient’s LDL-C [or systolic blood pressure] to within 10 points of goal is sufficient for reducing risk of future cardiac events”), (2) 4 items on urgency of getting to each goal (eg, “Lowering LDL-C [or systolic blood pressure] is often of secondary importance compared to other health conditions that many patients have”), and (3) 5 items on physician self-efficacy in treating patients to goal (eg, “Most of what is needed to get patients to their LDL-C [or systolic blood pressure] goal is outside of my control”). In addition, a subscale from a second validated instrument, the Preventive Medicine Attitudes and Activities Questionnaire,19,20 provided 3 items that measured perceived effectiveness at changing risk factor-related patient behavior (eg, “How effective do you think you are in changing your patients’ behavior with respect to a health diet?”). A 5-point Likert scale was used for all items, with higher scores indicating responses consistent with greater CVP-related attitudes and self-efficacy.
Talkin' 'bout synthetic generation
2006, Preventive MedicineA feasibility study of teaching motivational interviewing in a fully online environment using a virtual client
2022, International Journal of Health Promotion and Education