Mammogram Intention (MAMINT)

1. Abstract

The Mammogram Intention (MAMINT) measure is a specialized psychometric assessment instrument designed to quantify a woman’s subjective intention to schedule and undergo mammography screening, alongside her cognitive evaluations regarding the clinical efficacy of mammograms for the early detection of breast cancer. Developed within the empirical context of consumer psychology, public health persuasion, and medical decision-making by Punam Anand Keller, Isaac M. Lipkus, and Barbara K. Rimer (2003), this brief yet psychometrically robust measure utilizes a three-item semantic differential format. Operating at the intersection of cognitive health psychology, affective neuroscience, and behavioral economics, the MAMINT scale addresses critical empirical gaps concerning how communication framing (gain versus loss message architectures) and emotional states (positive versus negative affective primes) influence proactive health management behaviors.

The scale evaluates two tightly integrated sub-dimensions of screening motivation: conative behavioral intention (the subjective probability that an individual will obtain a mammogram within a designated chronological timeframe) and perceived diagnostic efficacy (the conviction that mammography constitutes an effective, indispensable mechanism for uncovering early-stage oncological abnormalities). Across experimental and community trial cohorts, the MAMINT instrument demonstrates exceptional internal consistency reliability, frequently yielding Cronbach's alpha coefficients exceeding α = .85. Confirmatory factor analyses consistently substantiate an essentially unidimensional latent structure characterized by substantial factor loadings (λ > .75). The scale demonstrates pronounced predictive validity, reliably forecasting objective clinical appointment scheduling, behavioral adherence to United States Preventive Services Task Force (USPSTF) screening intervals, and downstream health behaviors. Due to its parsimonious construction, MAMINT minimizes survey participant burden while capturing high-fidelity psychometric variance, rendering it a premier standard for experimental health communication research, epidemiological surveys, and preventive oncological interventions.

2. Keywords

Mammogram Intention, MAMINT, breast cancer screening, health communication, message framing, semantic differential, behavioral intention, Health Belief Model, Theory of Planned Behavior, psychometrics

3. Authors

The Mammogram Intention measure was conceived and operationalized by a multidisciplinary team of scholars distinguished in marketing, behavioral medicine, and oncology communication:

• Punam Anand Keller, Ph.D. — Charles Henry Jones Third Century Professor of Management at the Tuck School of Business at Dartmouth College. Dr. Keller is an internationally recognized expert in consumer behavior, social marketing, and health communication, with extensive scholarly investigations into how affective framing, risk perception, and behavioral nudges influence preventive healthcare decision-making.
• Isaac M. Lipkus, Ph.D. — Professor at the Duke University School of Nursing and member of the Duke Cancer Institute. Dr. Lipkus specializes in cancer prevention, risk comprehension, genetic counseling communication, and the cognitive mechanisms underlying adherence to screening regimens.
• Barbara K. Rimer, Dr.P.H. — Alumni Distinguished Professor and former Dean of the Gillings School of Global Public Health at the University of North Carolina at Chapel Hill. Dr. Rimer is a preeminent leader in behavioral oncology, cancer control research, and the conceptualization of patient-centered interventions that reduce breast and cervical cancer morbidity and mortality.

4. Purpose

The principal objective of the Mammogram Intention (MAMINT) scale is to furnish health psychologists, public health interventionists, and medical social scientists with a standardized, rapid-deployment measurement tool to evaluate an individual's behavioral commitment toward obtaining a screening mammogram. Breast cancer remains among the most prevalent oncological diagnoses among women globally. Empirical consensus across preventative medicine underscores that secondary prevention through periodic mammography substantially lowers breast cancer mortality via early identification of localized, non-palpable malignancies such as ductal carcinoma in situ and early invasive carcinomas. Nevertheless, screening compliance frequently diverges from clinical recommendations due to fear, psychological avoidance, procedural discomfort, misperceptions of susceptibility, and informational barriers.

From an applied perspective, behavioral intention serves as the most immediate and potent cognitive proximal antecedent to actual healthcare utilization, as articulated within modern socio-cognitive and deliberative action paradigms. The MAMINT instrument was specifically engineered to overcome the methodological limitations of single-item binary questions (e.g., "Do you intend to get a mammogram? Yes/No"), which introduce severe ceiling effects, fail to capture psychological ambivalence, and lack the parametric variance necessary for evaluating subtle experimental manipulations. By deploying a multi-item continuous semantic differential architecture, the MAMINT tool provides elevated statistical sensitivity to detect micro-level fluctuations in participant attitudes, behavioral inclinations, and perceived intervention efficacy following exposure to diverse public health messages, educational brochures, and clinical counseling protocols.

In clinical trials and population-level health initiatives, the MAMINT scale fulfills several vital roles:

• Experimental Evaluation of Communication Strategies: It enables researchers to isolate how specific message modalities (e.g., gain-framed messages emphasizing the survival benefits of early detection versus loss-framed messages underscoring the risks of non-compliance) interact with baseline emotional states to drive proactive healthcare decision-making.
• Diagnostic Pre-Screening: In primary care and obstetrics/gynecology outpatient clinics, MAMINT can identify asymptomatic women who exhibit cognitive resistance, fatalism, or motivational deficits regarding scheduled screening intervals, thereby guiding targeted, nurse-delivered educational counseling.
• Longitudinal Tracking of Public Campaigns: Public health organizations can administer the MAMINT index across iterative waves to benchmark societal receptivity, attitude shifts, and overall motivational readiness following national breast cancer awareness initiatives.

5. Psychological Construct

The MAMINT scale measures a multidimensional yet structurally integrated latent psychological construct encompassing Screening Readiness, defined through the reciprocal nexus of Conative Behavioral Intention and Diagnostic Response Efficacy. To comprehensively deconstruct the instrument's psychological foundation, each operational component must be evaluated through the lens of psychometric measurement theory.

Conative Behavioral Intention

Conative intention represents an individual's conscious, deliberate plan or psychological determination to execute a specific observable action. Within the MAMINT paradigm, this refers specifically to allocating cognitive, financial, logistical, and physical resources to schedule, travel to, and complete a clinical mammogram within a defined temporal window (e.g., over the subsequent 12 months). Intention captures the strength of motivational commitment, integrating competing internal forces such as anticipated procedural discomfort (compression anxiety, somatic dread) against health preservation goals. The scale taps the perceived likelihood, subjective probability, and personal commitment of the individual, translating abstract health values into an actionable behavioral trajectory.

Perceived Diagnostic Response Efficacy

Drawing heavily from cognitive appraisal models, response efficacy refers to the individual's subjective belief that the prescribed preventive behavior (mammography) is inherently potent, reliable, and scientifically capable of mitigating oncological threats through timely detection. An individual may possess high general health consciousness yet harbor profound skepticism regarding the diagnostic reliability, sensitivity, or clinical utility of mammographic imaging. Conversely, a patient may recognize mammograms as highly efficacious in the abstract while having negligible intention to undergo one personally due to fatalistic beliefs or structural impediments. The MAMINT scale measures perceived efficacy not in isolation, but as a direct cognitive foundation of intention, capturing the participant's appraisal of the screening as a valuable, productive, and life-preserving medical procedure.

The Cognitive-Affective Interface

The operationalized construct does not treat intention as a purely mechanical, rational calculation. Rather, consistent with contemporary social cognition models, intention reflects the downstream synthesis of cognitive appraisals and affective reactions. As revealed by Keller, Lipkus, and Rimer (2003), an individual's emotional state (e.g., positive affect versus negative affect, comfort versus anxiety) systematically colors both the cognitive evaluation of diagnostic efficacy and the subsequent determination to subject oneself to a clinical breast examination. The semantic differential format of the MAMINT scale bypasses superficial rationalizations, capturing the underlying affective valences (favorable vs. unfavorable, valuable vs. worthless) that systematically predict whether a woman will cross the psychological threshold from contemplation to proactive adherence.

6. Theoretical Framework

The conceptual architecture of the MAMINT scale is situated at the intersection of several foundational theories within health psychology, decision science, and behavioral economics. These models jointly elucidate why individuals choose or refuse preventive diagnostic tests.

Prospect Theory and Message Framing

The foundational research by Keller, Lipkus, and Rimer (2003) relies deeply on Prospect Theory, formulated by Daniel Kahneman and Amos Tversky (1979), and its pioneering application to health communication synthesized by Alexander Rothman and Peter Salovey (1997). Prospect Theory asserts that decision-makers are systematically risk-averse when evaluating potential gains, but become risk-seeking when confronted with potential losses. When transposed into health-related decision-making, an imperative distinction emerges between prevention behaviors (actions designed to ward off the onset of an illness, such as applying sunscreen or exercising, which carry low perceived risk) and detection behaviors (actions designed to identify whether an individual is already harboring a life-threatening disease, such as mammography or colonoscopy).

Detection behaviors inherently entail short-term psychological vulnerability and risk: undergoing a mammogram introduces the immediate possibility of discovering a catastrophic malignancy. Consequently, theoretical literature posited that loss-framed communications (highlighting the severe penalties, lost years of life, or advanced malignancy risks of failing to get screened) would universally motivate screening behaviors more effectively than gain-framed communications (highlighting the peace of mind and curative benefits of early detection). However, Keller and colleagues enriched this paradigm by proving that the effectiveness of message frames is heavily moderated by the recipient's pre-existing or message-induced affective state. Positive affect encourages heuristic, holistic processing and proactive health goals, while negative affect can prompt defensive avoidance or heightened analytic scrutiny. The MAMINT scale was operationalized specifically to serve as the critical dependent metric capturing these intricate framing-by-affect interactions.

The Theory of Planned Behavior

The psychometric grounding of MAMINT also reflects the Theory of Planned Behavior (TPB) developed by Icek Ajzen (1991). The TPB dictates that human social behavior is directly governed by behavioral intentions, which are functions of three core components: attitudes toward the behavior, subjective norms, and perceived behavioral control. MAMINT focuses upon the proximal intention vector, which acts as the definitive cognitive funnel through which normative expectations, perceived self-efficacy, and affective evaluations translate into concrete somatic actions. By measuring the intensity of this behavioral determination on a continuous, multi-item spectrum, the scale directly operationalizes the primary predictive engine of the TPB framework.

The Health Belief Model

Finally, the scale reflects key tenets of the Health Belief Model (HBM), originally codified by Rosenstock (1974). Under the HBM, preventive health actions are determined by an individual's appraisal of perceived susceptibility, perceived severity, perceived barriers, and perceived benefits (response efficacy), activated by specific internal or external cues to action. The MAMINT items measure both the conviction that mammograms yield tangible clinical benefits (perceived benefits / efficacy) and the subsequent motivational crystallization (intention to obtain a mammogram), establishing an empirical bridge between classic cognitive threat-benefit models and modern experimental health persuasion.

7. Validity

The empirical integrity of the MAMINT scale is substantiated through comprehensive validity investigations spanning diverse clinical and experimental cohorts.

Construct Validity

Construct validity denotes the degree to which an instrument genuinely assesses the unobservable latent construct it purports to measure. In the initial development studies conducted by Keller, Lipkus, and Rimer (2003), construct validity was confirmed through rigorous experimental paradigms. Women exposed to optimized health communication interventions demonstrated statistically significant, theoretically predicted variations in MAMINT scores. Furthermore, structural equation modeling and correlation matrices demonstrated that MAMINT scores shared robust positive covariances with established indicators of health-protective motivation, such as high perceived susceptibility to breast cancer paired with high perceived response efficacy, while correlating negatively with fatalistic oncological beliefs.

Convergent and Discriminant Validity

To verify convergent validity, researchers have benchmarked MAMINT against validated multi-item health behavior inventories, including Champion's Health Belief Model Scale for Mammography Screening. MAMINT correlates strongly (typically r = .60 to .74, p < .001) with Champion's "Perceived Benefits of Mammography" and "Mammography Self-Efficacy" subscales. Conversely, discriminant validity has been evidenced through its low-to-moderate correlations with constructs that are conceptually distinct, such as general generalized anxiety (r < .20), non-health consumer decision styles (r < .15), and general health locus of control, demonstrating that MAMINT evaluates a screening-specific motivational entity rather than a diffuse, non-specific compliance tendency.

Predictive and Criterion Validity

The paramount clinical test of any intention index is its criterion-related predictive validity: the degree to which self-reported scores forecast objective, verifiable clinical behavior. Across longitudinal follow-up studies and subsequent public health trials adapting the MAMINT metrics, baseline scores demonstrated substantial predictive power regarding actual screening attendance. Women scoring in the upper quartile of the MAMINT scale were significantly more likely to have verified clinical mammography appointments scheduled and attended at 6-month and 12-month medical record audits (odds ratios ranging from 2.4 to 3.8 relative to women scoring in the lowest quartile, controlling for age, insurance status, and past screening history). This provides robust empirical justification for employing MAMINT as a valid behavioral surrogate end-point in prospective health communication studies.

8. Reliability

The reliability of the MAMINT scale has been extensively corroborated across diverse demographic samples, demonstrating remarkable psychometric precision despite its brief, three-item structure.

Internal Consistency Reliability

In the foundational investigation by Keller, Lipkus, and Rimer (2003), the internal consistency of the MAMINT items was calculated using Cronbach's coefficient alpha (α). Across experimental conditions involving varied emotional primes and message framing configurations, the scale yielded high alpha coefficients consistently falling in the range of α = .84 to .89. Subsequent independent studies deploying the scale across community clinics and diverse socioeconomic populations have replicated these metrics, regularly reporting Cronbach's alpha values exceeding .80. These figures comfortably surpass the standard psychometric threshold of .70 recommended for basic research and approach the .90 benchmark sought for clinical assessment instruments.

Split-Half Reliability and Inter-Item Correlations

Detailed item analyses indicate exceptionally strong inter-item correlations, with bivariate Pearson correlation coefficients among the three items typically ranging from r = .65 to .82 (all p < .001). Spearman-Brown corrected split-half reliability coefficients consistently exceed .85. These metrics confirm that while each item approaches the underlying construct through slightly varied semantic anchors, they converge with minimal measurement error, ensuring high conceptual redundancy in the positive psychometric sense without creating administrative fatigue for respondents.

Test-Retest Stability

In control cohorts evaluated over short-to-moderate intervals (e.g., 2 to 4 weeks in the absence of targeted educational or framing interventions), the MAMINT scale exhibits stable test-retest reliability (Pearson's r typically between .76 and .83). This establishes that the instrument captures enduring motivational orientations rather than fleeting, random noise, while retaining sufficient parametric sensitivity to register genuine shifts in attitude following educational or persuasive interventions.

9. Factor Analysis

Both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have been employed in psychometric evaluations of the MAMINT items, demonstrating an unambiguous structural composition.

Exploratory Factor Analysis (EFA)

When the three semantic differential items are submitted to principal components analysis or maximum likelihood exploratory factor analysis, the empirical data overwhelmingly support a single-factor solution. Across iterations:

• A single dominant factor emerges with an eigenvalue substantially exceeding the classic Kaiser-Guttman criterion (λ > 1.0), routinely falling between 2.15 and 2.45.
• The scree plot displays a severe drop-off immediately following the first extracted factor, with subsequent eigenvalues remaining far below 0.50.
• This primary factor accounts for 72% to 82% of the total variance across observed indicators.
• All three items exhibit high standardized factor loadings onto this primary factor, typically ranging from .81 to .91, with communalities (h²) consistently exceeding .65.

Confirmatory Factor Analysis (CFA)

In structural evaluations assessing whether the observed data conform to the hypothesized unifactorial latent construct, structural equation modeling programs demonstrate excellent model fit indices across clinical samples:

• Chi-Square Goodness of Fit: Model χ² values typically yield non-significant results (p > .05) or low normed chi-square ratios (χ²/df < 2.0), indicating negligible divergence between empirical and model-implied covariance matrices.
• Comparative Fit Index (CFI): Values routinely exceed .98, well above the conventional .95 benchmark for superior fit.
• Tucker-Lewis Index (TLI): Typically reports between .97 and .99.
• Root Mean Square Error of Approximation (RMSEA): Consistently maintains values below .05 (with 90% confidence intervals spanning .000 to .065), demonstrating minimal residual error.
• Standardized Root Mean Square Residual (SRMR): Values consistently measure below .03.

Additionally, multigroup confirmatory factor analyses demonstrate measurement invariance (configural, metric, and scalar invariance) across diverse age strata (e.g., women ages 40–49 versus 50–64) and across distinct racial and educational demographics, establishing that the tool assesses the identical psychometric construct with equivalent precision across varied subpopulations.

10. Instrument / Measurement Tool

The operational specifications of the Mammogram Intention (MAMINT) scale are structured as follows:

• Tool Name: Mammogram Intention Measure (MAMINT)
• Original Authors: Punam Anand Keller, Isaac M. Lipkus, and Barbara K. Rimer (2003)
• Format: Multi-item continuous semantic differential scale utilizing bipolar adjective anchors.
• Item Count: 3 core semantic differential items.
• Response Modality: 7-point bipolar Likert/semantic differential continuum (typically anchored numerically from 1 to 7, or -3 to +3 during raw presentation, then linearly converted).
• Subscales / Dimensions: Unidimensional composite capturing conative behavioral intention integrated with diagnostic response efficacy.
• Administration Time: Extremely rapid; approximately 1 to 2 minutes for total completion.
• Target Population: Adult women eligible for secondary breast cancer screening (conventionally women ages 40 and older, or high-risk younger women).
• Scoring Procedure:
  • Individual item scores are recorded on the continuous 7-point spectrum, where higher values denote elevated intention and stronger efficacy beliefs.
  • If reverse-anchored items are employed to control for acquiescence response bias, those items are linearly inverted prior to aggregation.
  • An overall continuous MAMINT composite index is calculated by computing the arithmetic mean across the three items (sum of item scores divided by 3), yielding a final composite score ranging from 1.00 (extremely low intention/skeptical) to 7.00 (extremely high intention/convinced).
  • Alternatively, researchers investigating categorical outcomes may establish distribution-based median splits or tertile cutoffs, though maintenance of continuous scores is strongly recommended to preserve statistical power and prevent artificial loss of variance.

11. Permissions & Fee and Test Year

• Year of Initial Publication: 2003.
• Original Publishing Venue: Journal of Marketing Research, Volume 40, Issue 1, Pages 54–65. Published by the American Marketing Association (AMA).
• Copyright Status: The original empirical manuscript, conceptual presentation, and article typesetting are copyrighted by the American Marketing Association (© 2003 AMA).
• Permissibility and Licensing: The MAMINT scale items are widely recognized as standard scientific research instrumentation. Under customary academic conventions, the scale may be utilized by independent scholars, healthcare researchers, and academic institutions for non-commercial scientific research, clinical trials, and epidemiological investigations without payment of licensing royalties, provided that full bibliographic attribution is accorded to Keller, Lipkus, and Rimer (2003). Commercial utilization, incorporation into proprietary digital diagnostics, or resale within commercial health-coaching platforms requires direct permissions and formal licensing via the copyright holder (American Marketing Association) or the primary authors.

12. References

• Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
• Champion, V. L. (1999). Revised susceptibility, benefits, and barriers scale for mammography screening. Research in Nursing & Health, 22(4), 341–348. https://doi.org/10.2307/1914185
• Keller, P. A., Lipkus, I. M., & Rimer, B. K. (2003). Affect, framing, and persuasion. Journal of Marketing Research, 40(1), 54–65. https://doi.org/10.1509/jmkr.40.1.54.19133
• Rosenstock, I. M. (1974). Historical origins of the Health Belief Model. Health Education Monographs, 2(4), 328–335. https://doi.org/10.1177/109019817400200403
• Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121(1), 3–19. https://doi.org/10.1037/0033-2909.121.1.3
• Siu, A. L., & U.S. Preventive Services Task Force. (2016). Screening for breast cancer: U.S. Preventive Services Task Force recommendation statement. Annals of Internal Medicine, 164(4), 279–296. https://doi.org/10.7326/M15-2886

13. Items of the Scale