Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Practical and Relevant Self-Report Measures of Patient Health Behaviors for Primary Care Research

Practical and Relevant Self-Report Measures of Patient Health Behaviors for Primary Care Research Practical and Relevant Self-Report Measures of Patient Health Behaviors for Primary Care Research Russell E. Glasgow, PhD ABSTRACT Marcia G. Ory, PhD, MPH PURPOSE With increasing evidence for the value of behavior change counsel- ing, there is a need for health behavior measurements that can be implemented Lisa M. Klesges, PhD in primary care research. This article discusses criteria for and reviews self-report Maribel Cifuentes, RN measures to briefl y assess cigarette smoking, eating patterns, physical activity, and 4 risky drinking across the life course. It then proposes pragmatic measures for use Douglas H. Fernald, MA in practice-based research. Larry A. Green, MD METHODS Drawing from literature reviews, previous multisite studies, personal Kaiser Permanente, Denver, Colo communications with experts in the fi eld, and guidance from an expert panel, Texas A&M University System, College we identifi ed self-report behavior change measures and gave priority to items Station, Tex that addressed Healthy People 2010 goals, as well as those that were practical (ie, shorter, and easier to score and use for intervention), were sensitive to change, Mayo Clinic, College of Medicine, and produced results that could directly inform primary care intervention. Rochester, Minn Prescription for Health, University RESULTS Separate recommendations are described for measures for adults and of Colorado Health Sciences Center, for children/adolescents. We recommend a set of 22 items for adults and 16 items School of Medicine, Denver, Colo for adolescents to track succinctly their status on the 4 health behaviors above. Perfected measures remain elusive: newly developed measures of physical activ- ity and eating patterns are recommended, and in general, the brief measures for adults are currently better validated than are the child measures. CONCLUSIONS A set of totally satisfactory practical instruments for measuring behavior change in primary care settings does not yet exist. There is suffi cient progress to encourage use of and further research on the proposed items. Use of a common set of items across different interventions and projects will help to advance clinical and behavioral research in primary care settings. Ann Fam Med 2005;3:73-81. DOI: 10.1370/afm.261. INTRODUCTION ealth behavior counseling in primary care settings holds great 1-3 potential for disease prevention and management. Emerging Hevidence points to the importance of the physician as a change agent, the effi cacy of health behavior change interventions in primary care settings, and the value of behavioral change interventions for high-risk populations. It is timely to emphasize health behavior counseling in primary care Confl icts of interest: none reported given the large potential impact of personal behavior change on premature 1,7,8 death and avoidable suffering. The underperformance of the US health CORRESPONDING AUTHOR 9 care system is acknowledged and accompanied by keen interest to improve. Russell E. Glasgow, PhD The 3 primary care medical specialties, as well as primary care nurse practi- Clinical Research Unit tioners and physician’s assistants, exist in suffi cient numbers to make feasible Kaiser Permanente Colorado 10-12 greater attention to the health needs of primary care patients. Also, all 335 Road Runner Lane 3 of the primary care physician specialties—pediatrics, family medicine, Penrose, CO 81240 russg@ris.net and general internal medicine—are now calling for revisions in training ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 73 PRACTICAL HEALTH BEHAVIOR MEASURES and practice congruent with a greater emphasis on the that could be implemented in primary care research 13,14 health behaviors of their patients. without adding undue burden to patients or clinicians; Evaluation of behavioral interventions in primary and (3) discuss implementation, conditions of adminis- care research requires appropriate assessments of tar- tration, and directions for future research. geted health behaviors. Valid assessment methods are needed to identify patients who could benefi t METHODS from interventions, to inform the design of interven- tion strategies, and to evaluate intervention success. Context Whereas many instruments are available to assess To understand the current state of measurement in pri- health behaviors, few have been designed for or used mary care studies, we reviewed assessments proposed in primary care settings. by grantees in the fi rst round of P4H funding. Among Challenges for primary care settings include the the 17 sites, 15 focused on physical activity, 7 on risky “tyranny of competing demands.” Lengthy mea- drinking, 11 on cigarette smoking, and 14 on eating sures are not realistic to administer in practice-based patterns. All projects addressed at least 2 health behav- research. Practical measures are needed that are iors. Despite the common health behaviors addressed, broadly applicable, able to be self-administered, and there were few commonalities in the instruments appropriate for a range of age and cultural groups, and selected to measure health behaviors. The initial P4H that have suffi cient validity and reliability. Also, mea- projects proposed the following behavioral measures. sures should be congruent with national public health goals and sensitive to intervention effects, because the Physical Activity goal is to produce improvements in individual health as Physical activity logs, recalls, weekly assessments, and well as affect population health. screening questions were all used. Referenced question- Measures for primary care research need to be brief naires or sources included Physical Activity Scale for the 22,23 24 because of the lack of time or supervision to complete Elderly (PASE), PACE project assessment, Summary 25 3 lengthy instruments. Additionally, because primary of Diabetes Self-Care Activities, Healthy People 2010, 18,19 26 care is comprehensive —concerned with improving National Health Interview Survey (NHIS), Behavioral a variety of risk factors and conditions—focusing on Risk Factor Surveillance System—BRFSS, and Guide- 1 or 2 areas should not result in worse care in other lines for Adolescent Prevention Services (GAPS). areas. Unfortunately, almost all “gold standard” assess- ment procedures have focused on 1 health behavior Risky Drinking in isolation and attain part of their precision by being Number of alcoholic drinks per week, quantity of relatively lengthy, burdensome, or expensive. drinks, social aspects of drinking, and alcohol exposure In this article, we focus on assessment of 4 health and risk (adolescents only) assessments were all used. behaviors that collectively account for the majority Referenced questionnaires or sources included the Five of illness and mortality in the United States: cigarette Shot questionnaire, Alcohol Use Disorders Identifi ca- 30,31 26 27 28 smoking, eating patterns, physical inactivity, and risky tion Test (AUDIT), NHIS, BRFSS, and GAPS. 8,20 drinking. This work was conducted as part of Pre- scription for Health (P4H), a program of the Robert Cigarette Smoking Wood Johnson Foundation and the Agency for Health- Questions included whether the patient ever smoked, care Research and Quality. In its fi rst round, P4H was a current smoker, number cigarettes smoked and funded 17 practice-based research networks to fi eld quit attempts. Referenced questionnaires or sources test evidence-based strategies to improve the delivery, included Summary of Diabetes Self-Care Activities, 26 27 28 feasibility, and reach of behavior change counseling for NHIS, BRFSS, and GAPS. these 4 health behaviors in primary care practices. A major goal of the second round is to assess the extent Eating Patterns to which interventions are effective in attaining patient Food diaries, logs, 24-hour dietary recall, and nutrition behavior change. A common set of brief measures is intake questionnaires were used. Referenced sources necessary to compare outcomes across settings, inter- included the Summary of Diabetes Self-Care Activities, 32 33 ventions, and populations. Block Brief 2000 Questionnaire, “Rate Your Plate,” 26 27 3 28 Our purposes in this article are to (1) discuss issues NHIS, BRFSS, Healthy People 2010, and GAPS. in selecting measures of physical activity, risky drink- The length of the proposed assessment instruments ing, cigarette smoking, and eating patterns that are was surprising. Subsequent conversations with P4H appropriate for adult or pediatric primary care research; grantees suggested that it was diffi cult to implement (2) identify specifi c, practical behavioral assessments these measurement tools in practice. Although many ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 74 PRACTICAL HEALTH BEHAVIOR MEASURES (NIH) Web sites, using the Table 1. Criteria Used to Select Health Behavior Change Assessments search terms “nutrition,” “diet,” “eating,” “smoking,” “cigarette Criteria Rationale/Description smoking,” “alcohol consumption,” Primary “risky drinking,” “exercise,” and 1. Practicality A composite criterion including length; ease of administration, scor- ing, and interpretation; appropriateness to a wide range of set- “physical activity” with “self- tings and populations; and in the public domain, to reduce costs report,” “measurement,” “primary 2. Relationship to clinical Assessment of success toward achieving quantifi able public health care,” and “assessment.” In partic- and public health goals goals linked to health behavior change objectives. For example, preference was given to results that could be linked to achieve- ular, reviews from 2 major NIH- ment of Healthy People 2010 goals funded health behavior change 3. Validity Agreement with reference standard measures. We gave moderate consortia, the Behavior Change emphasis to convergent and divergent validity (correlations with other established measures of the same behavior and lack of cor- 34 Consortium and the Health relation with other constructs) Maintenance Consortium (http:// 4. Sensitivity to change Ability of a measure to detect improvements resulting from effective interventions was given the largest weight. To achieve sensitivity hmcrc.srph.tamu.edu/default. to change, an instrument must demonstrate other characteristics, aspx), provided excellent infor- such as test-retest reliability. The primary purpose of including behavioral measures was to evaluate whether signifi cant improve- mation on potential instruments. ments are produced An initial subset of measures 5. Relation to behavioral Degree to which instruments were behaviorally based and thus 21 was identifi ed by the fi rst 3 interventions directly related to P4H interventions (eg, it is less valuable to have a recall that produces detailed dietary nutrients, than mea- authors, with exclusion of mea- sures that assess changes in eating behavior) sures that were longer than 10 Secondary items or that were not appropri- 6. Cultural and literacy Extent to which measures were appropriate for and had been trans- ate for self-administration. Can- issues lated into and validated in other languages, especially Spanish didate measures were initially 7. Norms and breadth Having data on national or regional norms. Data from instruments of use with well-established norms allow comparison of interventions reviewed to establish their suit- and populations to larger reference groups ability, and tables were devel- 8. Age appropriateness Appropriate across the full age range and functional levels. Given special attention for children and adolescents oped that outlined information 9. Reliability Given the brevity required, we did not place strong emphasis on on the various selection criteria. internal consistency (α coeffi cients), but gave considerable weight Summary information and to test-retest reliability tables were presented to a con- 10. Usability for feedback Scales that can be scored immediately or can identify areas for and interventions improvement to help with patient goal setting sensus panel of P4H program representatives and invited of the proposed instruments had demonstrated validity experts. The panel included 12 experts in the areas of for screening, their ability to detect changes in health primary care, practice-based research, and quantitative behaviors is uncertain. and qualitative analyses. The panel agreed that priority for further recommendations should be based on assess- Procedures and Criteria Used ments demonstrated to have (1) sensitivity to change in Making Recommendations or intervention effects; (2) broad applicability, espe- To identify practical measures for assessing and track- cially to minority populations or validated in Spanish; ing health behaviors, we needed to specify selection (3) normative data; and (4) usefulness for intervention criteria. Greatest weight was given to 5 primary crite- planning. Final revised recommendations were based on ria: practicality, clinical and public health relevance, discussion by this consensus panel. validity, sensitivity to change, and relation to behav- The following key issues emerged from the panel ioral intervention. The 5 secondary criteria listed at the discussion: (1) the number of items needed to be even bottom of Table 1 were also considered. Because child briefer than originally proposed if the expectation was and adolescent risk behaviors differ in frequency and that all practices would assess multiple behaviors; (2) nature from those of adults, separate assessments were the wording of several items did not apply to primary generally necessary. care; (3) response formats of several items needed to be Candidate measures were identifi ed from several changed to fi t time frames relevant to practice-based sources, including instruments implemented in the fi rst research; (4) a recommended list of brief items should phase of P4H, review of published literature, sugges- be provided, as well as a secondary recommendation tions of professional contacts (because many instruments where warranted, of slightly longer instruments if they are under development) and consensus statements and had superior qualities; and (5) sensitivity to change was guidelines for recommended assessments. Searches were very important, and several widely used screening mea- conducted on PubMed and National Institute of Health sures did not meet this criterion. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 75 PRACTICAL HEALTH BEHAVIOR MEASURES Figure 1. Adult Measures of Behavior Change Physical activity (RAPA) * Scoring – see Web site: http://hmcrc.srph.tamu.edu/Measurement%20folder/PA/RAPA%20fnl%201004.doc Physical activities are activities where you move and increase your heart rate above its resting rate, whether you do them for pleasure, work, or transportation. The following questions ask about the amount and intensity of physical activity you usually do. The intensity of the activity is related to the amount of energy you use doing these activities. Examples of physical activity intensity levels: Light activities Your heart beats slightly faster than normal You can talk and sing Light exercise Light stretching Light vacuuming or yard work Moderate activities Your heart beats faster than normal You can talk but not sing Brisk walking Aerobics class Strength training Swim gently Vigorous activities Your heart rate increases a lot You can’t talk, or your talking is broken up by large breaths Aerobics classes Jogging, Running, Singles tennis, or Power Walking Racquetball, Pickle ball How physically active are you? (answer Yes or No to each) Yes No I rarely or never do any physical activities. I do some light and/or moderate physical activities, but not every week. I do some light physical activity every week. I do moderate physical activity every week but less than 5 days per week or less than 30 minutes on those days. I do vigorous physical activities every week, but less than 3 days per week or less than 20 minutes on those days. I do 30 minutes or more per day of moderate physical activities 5 or more days per week. I do 20 minutes or more per day of vigorous physical activities 3 or more days per week. I do activities to increase muscle strength, such as lifting weights or calisthenics, once a week or more. I do activities to improve fl exibility, such as stretching or yoga, once a week or more. *Adapted with permission from “How Physically Active Are You?” from the University of Washington Health Promotion Research Center, Seattle, Wash. Figure 1 continues ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 76 PRACTICAL HEALTH BEHAVIOR MEASURES Risky drinking (BRFSS 2003 ) Scoring – Number drinks. Binge drinking = 5 or more drinks for men; 4 or more for women A drink of alcohol is defi ned as 1 can or bottle of beer, 1 glass of wine, 1 can or bottle of wine cooler, 1 cocktail, or 1 shot of liquor. During the past 30 days, how many days per week or per month did you have at least 1 drink of any alcoholic beverage? [if none, STOP] _______________ On the days when you drank, about how many drinks did you drink on average? ______________ Considering all types of alcoholic beverages, how many times during the past 30 days did you have 5 or more drinks on an occasion? (for women the threshold for binge drinking is 4 drinks). _________________ Cigarette smoking Scoring – any current smoking, especially daily smoking as trigger for intervention Have you smoked at least 100 cigarettes in your entire life? [if no, STOP] (BRFSS 2003) _______________ Have you smoked at least part of a cigarette in the last 7 days? [if no, STOP] (SRNT) _______________ During a typical 7-day period how many cigarettes did you smoke? (BCC) _______________ Eating patterns – (Starting the Conversation – Diet) Scoring — Contact Alice Ammerman at Alice_Ammerman@unc.edu (personal communication, May 15, 2004) How many times a week do you eat fast food meals or snacks? ___ 1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many servings of fruit or vegetables do you eat each day? ___ 1 serving ___ 2 servings ___3 or more servings ___ None ___ Don’t know/not sure How many regular sodas or glasses of sweet tea do you drink each day? (one glass is an 8 oz serving) ___<1 glass ___ 1 glass ___ 2 glasses ___ 3 glasses ____4 or more ___ None ___Don’t know/not sure How many times a week do you eat beans (like pinto or black beans), chicken or fi sh? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many times a week do you eat regular snack chips or crackers (not the low-fat)? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many times a week do you eat desserts and other sweets? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How much margarine, butter or meat fat do you use to season vegetables or put on potatoes, bread, or corn? ___ Very little ___ Some ___ A lot BRFSS = Behavioral Risk Factor Surveillance System; SRNT = Society for Research on Nicotine and Tobacco; BCC = Behavior Change Consortium. RESULTS was it possible to include entire instruments. Several For each target behavior, we offer both a primary of our recommendations endorsed new instruments, recommendation, and in most cases, a secondary rec- and potential users should review these measures for 25,27,34,36-45 ommendation, which often involves an elaboration of updates before implementing. Table 3 (avail- the primary measure that might be applicable for sites able online only at http://www.annfammed.org/ wishing to devote more time or priority to that behav- cgi/content/full/3/1/73/DC1) summarizes the ior. Our primary recommendations are summarized characteristics of recommended measures. 34-36 27,37-41 in Figure 1 and Table 2 (which are available online at http://www.annfammed.org/cgi/con- Recommended Measures for Adolescents tent/full/3/1/73/DC1); secondary recommenda- and Children 38,39,42-44,45 25,27,30,31,36 tions are in Appendixes 1 and 2 Only a few measures of health behaviors in children (which are available online at http://www.annfammed. or adolescents were located that fi t most of the desired org/cgi/content/full/3/1/73/DC1). For adults, 22 items criteria, and none were identifi ed that reported sen- are recommended as a minimum assessment across sitivity to change. Some instruments with acceptable all 4 behaviors. The corresponding number for ado- validity and reliability were identifi ed, but they were lescents, ages 12 to 17 years is 16, and 12 items for too long or cumbersome to be feasible in primary younger children. Although many of the items rec- care settings. Given the paucity of available measures, ommended are from validated instruments, seldom recommended items were sometimes based on items ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 77 PRACTICAL HEALTH BEHAVIOR MEASURES from adult samples or that had face validity but no Children. Risky drinking in children younger than established measurement properties in adolescents and 12 years is infrequent, and validated measures of sus- children. ceptibility are not currently available. The following measures are recommended for assessing adolescent and child behavior. Table 3 pres- Smoking ents the actual items. Adolescents. Two items from the Smoking Uptake Continuum assessment are recommended. These Physical Activity items assess susceptibility to smoking, with the fi rst Adolescents. Three items are recommended, including item asking about experimentation with cigarettes and 2 items from the PACE+ project and 1 item from the the second asking how many days, in the last 30, that Youth Risk Behavior Surveillance System (YRBSS). an adolescent smoked. The second item can be used to The PACE+ items, which assess frequency of physi- calculate attainment of Healthy People 2010 goals. cal activity in the last 7 days and in a typical week, The secondary recommendation for adolescent have been validated in a diverse population of middle- smoking is to include 8 additional items from the school adolescents. The items demonstrated accept- Smoking Uptake Continuum (Appendix 1). This 10- able measurement properties (sensitivity 71%, specifi c- item measure categorizes youth into 7 levels of smok- ity 63%; signifi cantly correlated with accelerometer ing behavior from committed never smoker to current readings r = .40, P<.001). In addition, our panel established smoker. The instrument is validated with believed it was important to assess sedentary behav- adolescents aged 12 to 18 years and has demonstrated ior, so an item from the YRBSS was recommended to good predictive validity (<6% committed never smok- evaluate hours of television watched on an average ers progress to established smokers over a 3- to 4-year school day. All recommended items can be used to period, susceptible never smokers are 2 to 3 times more address the Healthy People 2010 goals. A secondary likely to initiate smoking over a 3- to 4-year period recommendation was to assess sedentary behaviors of than are committed never smokers). computer and video game use via an adaptation of the Children. A primary recommendation is to assess YRBSS item that assesses television viewing. Measure- children’s susceptibility to future smoking based on 4 ment characteristics and normative data are not cur- items from Jackson’s Susceptibility to Smoking scale. rently available for this item. This measure has been validated in children in grades Younger Children. Obtaining reliable and valid 3 to 5 and found to be predictive of smoking initiation self-reported physical activity in younger children is 2 years after assessment. 46,47 problematic. No feasible and reliable self-report instruments were identifi ed for children, and objective Eating Patterns measures such as activity monitors were not considered Adolescents. The literature review identifi ed a few practical. Therefore, no recommendation is made for measures that assessed nutrient intake and had been this age-group. validated in clinical settings with adolescents and chil- 49,50 dren. These instruments were considered too long, Risky Drinking however, to be feasible in primary care settings. The Adolescents. The 3 items recommended are based on panel also emphasized the importance of assessing questions from the 2003 BRFSS. While our panel rec- dietary behaviors and food groups, such as fruits and ognized that overall prevalence of alcohol use would vegetables, rather than nutrients, for use in behavioral be lower in adolescents compared to adults, these items counseling. were recommended to identify current problems with We recommend an 8-item instrument developed alcohol consumption. They are also useful for calculat- by Ammerman and colleagues (Alice Ammerman, ing Healthy People 2010 goal attainment. personal communication, May 15, 2004) that assesses As a secondary assessment of risky drinking, 6 items dietary patterns, such as frequency of fast-food intake, from the CRAFFT screening test are recommended. fat snack intake, and sweetened beverage consump- This measure has been validated in clinical settings tion. The instrument was designed to be completed with diverse populations of adolescents aged 13 to19 by adolescents on their own and together with parents 42-44 years and has demonstrated reliability and validity. for younger children. It was designed using literature An adapted version of the instrument that screens review, expert input, and feasibility testing, and mea- only for alcohol, instead of alcohol and drug use, was surement properties are currently being evaluated. believed to be most practical. Normative data are avail- A secondary recommendation is a 2-item assessment able on the longer CRAFFT instrument, but cutoffs for of fruit and vegetable intake developed by the PACE+ specifi c alcohol-related problems are not. project. These items are consistent with Healthy ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 78 PRACTICAL HEALTH BEHAVIOR MEASURES People 2010 goals and have been validated in a diverse Time frames of many of the AUDIT items prevent pre- population of middle school-aged adolescents. They cise change estimates. have demonstrated acceptable measurement properties (signifi cant correlations with 3-day food record; sen- Cigarette Smoking sitivity 81%, specifi city 47%, positive predictive value Primary measure. Three items from national health sur- 27,34,36 79%). Although the items have not been validated in veys. These items (Figure 1) are recommended to younger children, the panel believed that younger chil- assess whether respondents ever smoked, their current dren could complete these items with assistance from a smoking status, and extent of smoking. Individual items parent or guardian. are taken from national surveys, and thus there are normative data; however, with the exception of the sec- Recommended Measures in Adults ond item, these are not standard measures from major Our review indicated that there are now several practi- trials to evaluate smoking interventions. There was cal lifestyle measures being tested for use in adults, 2 of concern about wording of some of the lengthier and which are primary recommendations. We attempted to more complex items used in smoking cessation trials. select instruments that could bridge the vast diversity The recommended items form an effi cient condensed of adult populations (Figure 1). set (if persons answer “no” to initial items, they skip the remaining questions) and allow tracking on Healthy Physical Activity People 2010 goals. Primary measure. The Rapid Assessment Physical Secondary measure. Six items recommended by 35 36 Activity Scale (RAPA). This scale involves 9 yes/no the Society for Research on Nicotine and Tobacco items assessing the type and amount of physical activ- (embedded within 3 questions in Appendix 2) to mea- ity in which adults engage. Clinicians and measurement sure cessation in cessation trials. These items refl ect experts at the University of Washington developed this expert consensus assessment of abstinence parameters scale for primary care settings. In preliminary validation (eg, temporal intervals, slips). They are recommended studies it compares well with longer scales. A major for those emphasizing smoking behaviors. This approach advantage is that it enables respondents to visualize dif- recognizes that there will be relapses and also allows ferences in activity intensity. Pictures and examples can survival analyses. be made culturally and setting relevant. Other advan- tages are a quick and easy scoring sheet, and availability Eating Patterns in different languages (eg, Spanish and Vietnamese). Primary measure. Starting the Conversation (STC) Secondary measure. Seven items from the BRFSS. – Diet. (Alice Ammerman, personal communication, The main advantage of the BRFSS is that it permits May 15, 2004) This 7-item scale comes out of a pro- comparison with national norms, and Spanish trans- gram of nutrition research by Ammerman and col- lations are available. Although separating types of leagues. The advantage of this scale is its simplicity activity may be helpful, the original purpose of the and relative brevity. It also assesses food patterns vs BRFSS was for surveillance, not intervention tracking. nutrient or fat intake and provides tailored strategies Additionally, responses are often ambiguous because for behavioral counseling. The validity of the STC of respondent failure to differentiate among intensity is currently being examined in the WISEWOMAN levels. We recommend RAPA over this scale, for likeli- Study, with comparisons to longer, more traditional hood of more accurate reporting, and the use of graph- nutritional assessments. ics to clarify different levels of activity. Both scales can Secondary measure. The Summary of Diabetes be mapped to Healthy People 2010 recommendations. Self-Care Activities (SDSCA). This scale includes 4 dietary items that address the most commonly targeted Risky Drinking dietary recommendations—eg, low fat, high fi ber, and Primary measure. Three items from the BRFSS. These increased fruit and vegetable intake. The questions cor- items offer a fast, simple way to detect risky drinking relate reasonably well with much longer reference stan- and to examine changes in binge drinking. Findings dard instruments. The SDSCA has been widely used in can be compared with national norms and related to various formats (paper and computer administered) and Healthy People 2010 goals. found to be sensitive to change in intervention studies 30,31 Secondary measure. AUDIT. This 10-item with diabetes patients, but has not been validated in instrument is often used in national and international nondiabetic study samples. The questions seem appli- studies to screen for risky drinking, and it is available cable to nondiabetic adults, but we recommend the in Spanish. It is less relevant for intervention studies STC items because they lead directly to intervention and assessing change over time than it is for screening. planning. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 79 PRACTICAL HEALTH BEHAVIOR MEASURES DISCUSSION approaches, such as item response theory, is needed to Our goal was to derive a set of practical, yet valid, determine whether it is possible to further streamline items that could evaluate patient behavior change in these assessments. the 4 priority areas across the diverse P4H projects. To read or post commentaries in response to this article, see it This same set of items should be appropriate for online at http://www.annfammed.org/cgi/content/full/3/1/73. most primary care-based research and practices in the United States. Given our mandate, the failure of tradi- Key words: Health behavior; risk factors; needs assessment; review; pri- tional review procedures to identify many appropriate mary health care measures, and the frequent lack of information on key criteria (eg, sensitivity to intervention; cultural appro- Submitted September 15, 2004; submitted, revised November 29, 2004; accepted December 1, 2004. priateness), we were forced to rely to a moderate extent upon informal networks of leading researchers and Funding support: This project was supported by the Robert Wood John- organizations, and upon subjective estimates of which son Foundation in supporting the Prescription for Health project. items best fulfi lled our criteria. Consistent use of the proposed measures would References greatly improve the science of health promotion among primary care patients. It is diffi cult to compare results 1. U. S. Preventive Services Task Force. Guide to Clinical Preventive Ser- vices. 2nd ed. Baltimore, MD: Williams & Wilkins; 1996. of studies from different interventions and populations 2. Centers for Disease Control and Prevention. The Guide to Community if each study uses measures of differing or unknown Preventive Services. 2002 Available at: http://www.thecommunity- reliability, validity, and sensitivity to change—even if guide.org. results are converted to standard units such as effect 3. U. S. Department of Health and Human Services. Healthy People size. We propose the items in Figure 1 and Table 2, 2010: Understanding and Improving Health. Washington, DC: Govern- as a relatively succinct set of items for use in primary ment Printing Offi ce; 2000. 52,53 care research and practical clinical trials. These 4. Ockene IS, Hebert JR, Ockene JK, et al. Effect of physician-delivered nutrition counseling training and an offi ce-support program on satu- recommendations should also be useful to clinicians rated fat intake, weight, and serum lipid measurements in a hyper- and health plans for identifi cation of health behavior lipidemic population: Worcester Area Trial for Counseling Hyperlipid- change needs and quality improvement efforts. emia (WATCH). Arch Intern Med. 1999;159:725-731. Additional research is needed on these measures. 5. Whitlock EP, Orleans CT, Pender N, Allan J. Evaluating primary care behavioral counseling interventions: an evidence-based approach. Am Some items are very new, and more data are needed J Prev Med. 2002;22:267-284. on their continued performance. Several items recom- 6. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the mended have not been fully tested in terms of tradi- incidence of type 2 diabetes with lifestyle intervention or metformin. tional psychometric standards. There is, however, a N Engl J Med. 2002;346:393-403. compelling need for a practical set of items appropriate 7. McGinnis JM, Foege WH. Actual causes of death in the United for assessing multiple health behaviors in primary care. States. JAMA. 1993;270:2207-2212. We posit that the recommended items are a better 8. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States. JAMA. 2004;291:1238-1245. solution to this need than recommending nothing, or a much longer, and less clinically useful set of mea- 9. Institute of Medicine, Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st sures that would be much less likely to be used. These Century. Washington, DC: National Academy Press; 2001. recommendations should be periodically reevaluated 10. Luire JD, Goodman DC, Wennberg JE. Benchmarking the future gen- and revised as new information becomes available. eralist workforce. Eff Clin Pract. 2002;5:58-66. This report is limited to assessments of these 4 health 11. Shipman LA, Lurie JD, Goodman DC. The general pediatrician: behaviors and does not address other important self- projection of future workforce supply and requirements. Pediatrics. 2004;113:435-442. management goals that may be set between patients and their health care team. We restricted our focus to 12. Council on Graduate Medicine Education. Physician Workforce Policy Guidelines for the U.S. 2000-2020. Albany, New York: Center for self-report measures and did not review direct observa- Health Workforce Studies, University of Albany; 2004. tion or mechanical instrumentation methods. 13. Larson EB, Fihn SD, Kirk LM, et al. The future of general internal We recommend 3 specifi c areas of future research. medicine. Report and recommendations from the Society of General First, these proposed items should be compared directly Internal Medicine (SGIM) Task Force on the Domain of General Inter- nal Medicine. J Gen Intern Med. 2004;19:69-77. to “gold standard” assessments. Second, more studies 14. Future of Family Medicine Project Leadership Committee. The future are needed of their sensitivity to change; acceptability of family medicine: a collaborative project of the family medicine in a spectrum of family medicine, internal medicine, community. Ann Fam Med. 2004;2:S3-S32. and pediatric practices; and results under different 15. Shumaker SA, Schron EB, Ockene JK, McBee WL. The Handbook of conditions of administration, in particular computer- Health Behavior Change. 2nd ed. New York, NY: Springer Publishing automated assessment. Finally, research using modern Company; 1998. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 80 PRACTICAL HEALTH BEHAVIOR MEASURES 16. Stange KC, Woolf SH, Gjeltema K. One minute for prevention: the 37. Prochaska JJ, Sallis JF, Long B. A physical activity screening measure power of leveraging to fulfi ll the promise of health behavior counsel- for use with adolescents in primary care. Arch Pediatr Adolesc Med. ing. Am J Prev Med. 2002;22:320-323. 2001;155:554-559. 17. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical 38. Centers for Disease Control and Prevention. National Center for Care Survey: 2001 summary. Adv Data. 2003;22:1-44. Chronic Disease Prevention and Health Promotion. Youth Risk Behav- ior Surveillance System. Available at: http://www.cdc.gov/Healthy 18. Starfi eld B. Primary care: Balancing Health Needs, Services, and Technol- Youth/yrbs/about_yrbss.htm. ogy. Cambridge: Oxford University Press; 1998. 39. Choi WS, Gilpin EA, Farkas AJ, Pierce JP. Determining the probability 19. Institute of Medicine. Primary Care: America’s Health in a New Era. of future smoking among adolescents. Addiction. 2001;2:313-323. Washington, D.C.: National Academy of Sciences; 1996. 40. North Carolina Department of Health and Human Services. Physical 20. Fine LJ, Philogene GS, Grambling R, Coups EJ, Sinha S. Prevalence activity and nutrition behaviors monitoring form. Available at: http:// of multiple behavioral risk factors in the United States: Results www.eatsmartmovemorenc.com/data/PAN%20Behaviors%20Monitori from the 2001 National Health Interview Survey. Am J Prev Med. ng%20Form.pdf 2004;27(2S):18-24. 41. Jackson C. Cognitive susceptibility to smoking and initiation 21. Robert Wood Johnson Foundation, Agency for Healthcare Research of smoking during childhood: a longitudinal study. Prev Med. and Quality. PrescriptionForHealth Web site. Available at: http://www. 1998;27:129-134. prescriptionforhealth.org. 42. Knight JR, Shrier LA, Bravender TD, Farrell M, Vander Bilt J, Shaffer 22. Washburn RA, Smith KW, Jette AM, Janney CA. The physical activity HJ. A new brief screen for adolescent substance abuse. Arch Pediatr scale for the elderly (PASE): development and evaluation. J Clin Epide- Adolesc Med. 1999;153:592-596. miol. 1993;46:153-162. 43. Knight JR, Sherritt L, Shrier La, Karris SK, Chang G. Validity of the 23. Washburn RA, McAuley E, Katula J, Mihalko SL, Boileau RA. The CRAFFT substance abuse screening test among adolescent clinic physical activity scale for the elderly (PASE): evidence for validity. J patients. Arch Pediatr Adolesc Med. 2002;156:607-614. Clin Epidemiol. 1999;52:643-651. 44. Cummings LH, Chan KK, Burns KM, Blume AW, Larimer M, Marlatt 24. Calfas KJ, Sallis JF, Zabinski MF, et al. Preliminary evaluation of GA. Validity of the CRAFFT in American-Indian and Alaska-Native multi-component program for nutrition and physical activity change adolescents: screening for drug and alcohol risk. J Stud Alcohol. in primary care: PACE+ for adults. Prev Med. 2002;34:153-161. 2003;64:727-732. 25. Toobert DJ, Hampson SE, Glasgow RE. The summary of Diabetes 45. Prochaska JJ, Sallis JF. Reliability and validity of a fruit and Self-Care Activities Measure: results from seven studies and revised vegetable screening measure for adolescents. J Adolesc Health. scale. Diabetes Care. 2000;23:943-950. 2004;34:163-165. 26. U. S. Department of Health and Human Services. National Health 46. Klesges LM, Baranowski T, Beech B, et al. Social desirability bias in Interview Survey (NHIS). 2004. Available at: http://www.cdc.gov/ self-reported dietary, physical activity and weight concerns measures nchs/nhis.htm. in 8- to 10-year old African-American girls: Results from the Girls 27. U. S. Department of Health and Human Services. Behavioral Risk Health Enrichment Multisite Studies (GEMS). Prev Med. 2004;38: Factor Surveillance System. 2004. Available at: http://www.cdc.gov/ S78-S87. brfss/index.htm. 47. Welk GJ, Corbin CB, Dale D. Measurement issues in the assessment 28. Elster A. The American Medical Association guidelines for adolescent of physical activity in children. Res Q Exerc Sport. 2000;71(2 Suppl): preventive services. Arch Pediatr Adolesc Med. 1997;151:958-959. S59-S73. 29. Seppa K, Lepisto J, Sillanaukee P. Five-shot questionnaire on heavy 48. Substance Abuse and Mental Health Services Administration. Results drinking. Alcohol Clin Exp Res. 1998;22:1788-1789. from the 2002 National Household Survey on Drug Use and Health: National Findings. Pediatrics. 2001. Rockville, MD. 30. The Alcohol Use Disorders Identifi cation Test: Guidelines for Use in Prima- ry Health Care. Geneva, Switzerland: World Health Organization, 1989. 49. Prochaska JJ, Sallis JF, Rupp J. Screening measure for assessing dietary fat intake among adolescents. Prev Med. 2001;33:699-706. 31. Saunders JB, Aasland OB, Babor TF, et al. Development of the Alcohol Use Disorders Identifi cation Test (AUDIT): WHO collaborative 50. Dennison BA, Jenkins PL, Rockwell HL. Development and valida- project on early detection of persons with harmful alcohol consump- tion of an instrument to assess child dietary fat intake. Prev Med. tion--II. Addiction. 1993; 88:791-804. 2000;31:214-224. 32. Block G, Gillespie C, Rosenbaum EH, Jenson C. A rapid food 51. Will JC, Farris RP, Sanders CG, Stockmyer CK, Finkelstein EA. Health screener to assess fat and fruit and vegetable intake. Am J Prev Med. promotion interventions for disadvantaged women: overview of the 2000;18:284-288. WISEWOMAN projects. J Womens Health. 2004;13:484-502. 33. Gans KM, Sundaram SG, McPhillips JB, Hixson ML, Linman Carlton 52. Tunis SR, Stryer DB, Clancey CM. Practical clinical trials. Increas- RA. Rate your plate: an eating pattern assessment and educational tool ing the value of clinical research for decision making in clinical and used at cholesterol screening programs. J Ntr Educ. 1993;25:29-36. health policy. JAMA. 2003;290:1624-1632. 34. Ory MG, Jordan PJ, Bazzare T. The Behavior Change Consortium: set- 53. Glasgow RE, Magid DJ, Beck A, Ritzwoller D, Estabrooks PA. Practi- ting the stage for a new century of health behavior change research. cal clinical trials for translating research to practice: design and mea- Health Educ Res. 2002;17:500-511. surement recommendations. Med Care. In press. 35. University of Washington Health Promotion Research Center. Rapid 54. Wagner EH. The role of patient care teams in chronic disease man- Assessment Physical Activity Scale (RAPA). Available at: http://hmcrc. agement. BMJ. 2000;320:569-572. srph.tau.edu. 55. Babor TF, Sciamanna CN, Pronk N. Assessing multiple risk behaviors 36. Society for Research on Nicotine and Tobacco Web site. Available in primary care: screening issues and related concepts. Am J Prev at http://hmcrc.srph.tamu.edu/Measurement%20folder/Smoking/ Med. 2004; 27(2S):42-53. Tobacco%20SNT.pdf. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Annals of Family Medicine Unpaywall

Practical and Relevant Self-Report Measures of Patient Health Behaviors for Primary Care Research

The Annals of Family MedicineJan 1, 2005

Loading next page...
 
/lp/unpaywall/practical-and-relevant-self-report-measures-of-patient-health-C00vhXy5lJ

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1544-1709
DOI
10.1370/afm.261
Publisher site
See Article on Publisher Site

Abstract

Practical and Relevant Self-Report Measures of Patient Health Behaviors for Primary Care Research Russell E. Glasgow, PhD ABSTRACT Marcia G. Ory, PhD, MPH PURPOSE With increasing evidence for the value of behavior change counsel- ing, there is a need for health behavior measurements that can be implemented Lisa M. Klesges, PhD in primary care research. This article discusses criteria for and reviews self-report Maribel Cifuentes, RN measures to briefl y assess cigarette smoking, eating patterns, physical activity, and 4 risky drinking across the life course. It then proposes pragmatic measures for use Douglas H. Fernald, MA in practice-based research. Larry A. Green, MD METHODS Drawing from literature reviews, previous multisite studies, personal Kaiser Permanente, Denver, Colo communications with experts in the fi eld, and guidance from an expert panel, Texas A&M University System, College we identifi ed self-report behavior change measures and gave priority to items Station, Tex that addressed Healthy People 2010 goals, as well as those that were practical (ie, shorter, and easier to score and use for intervention), were sensitive to change, Mayo Clinic, College of Medicine, and produced results that could directly inform primary care intervention. Rochester, Minn Prescription for Health, University RESULTS Separate recommendations are described for measures for adults and of Colorado Health Sciences Center, for children/adolescents. We recommend a set of 22 items for adults and 16 items School of Medicine, Denver, Colo for adolescents to track succinctly their status on the 4 health behaviors above. Perfected measures remain elusive: newly developed measures of physical activ- ity and eating patterns are recommended, and in general, the brief measures for adults are currently better validated than are the child measures. CONCLUSIONS A set of totally satisfactory practical instruments for measuring behavior change in primary care settings does not yet exist. There is suffi cient progress to encourage use of and further research on the proposed items. Use of a common set of items across different interventions and projects will help to advance clinical and behavioral research in primary care settings. Ann Fam Med 2005;3:73-81. DOI: 10.1370/afm.261. INTRODUCTION ealth behavior counseling in primary care settings holds great 1-3 potential for disease prevention and management. Emerging Hevidence points to the importance of the physician as a change agent, the effi cacy of health behavior change interventions in primary care settings, and the value of behavioral change interventions for high-risk populations. It is timely to emphasize health behavior counseling in primary care Confl icts of interest: none reported given the large potential impact of personal behavior change on premature 1,7,8 death and avoidable suffering. The underperformance of the US health CORRESPONDING AUTHOR 9 care system is acknowledged and accompanied by keen interest to improve. Russell E. Glasgow, PhD The 3 primary care medical specialties, as well as primary care nurse practi- Clinical Research Unit tioners and physician’s assistants, exist in suffi cient numbers to make feasible Kaiser Permanente Colorado 10-12 greater attention to the health needs of primary care patients. Also, all 335 Road Runner Lane 3 of the primary care physician specialties—pediatrics, family medicine, Penrose, CO 81240 russg@ris.net and general internal medicine—are now calling for revisions in training ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 73 PRACTICAL HEALTH BEHAVIOR MEASURES and practice congruent with a greater emphasis on the that could be implemented in primary care research 13,14 health behaviors of their patients. without adding undue burden to patients or clinicians; Evaluation of behavioral interventions in primary and (3) discuss implementation, conditions of adminis- care research requires appropriate assessments of tar- tration, and directions for future research. geted health behaviors. Valid assessment methods are needed to identify patients who could benefi t METHODS from interventions, to inform the design of interven- tion strategies, and to evaluate intervention success. Context Whereas many instruments are available to assess To understand the current state of measurement in pri- health behaviors, few have been designed for or used mary care studies, we reviewed assessments proposed in primary care settings. by grantees in the fi rst round of P4H funding. Among Challenges for primary care settings include the the 17 sites, 15 focused on physical activity, 7 on risky “tyranny of competing demands.” Lengthy mea- drinking, 11 on cigarette smoking, and 14 on eating sures are not realistic to administer in practice-based patterns. All projects addressed at least 2 health behav- research. Practical measures are needed that are iors. Despite the common health behaviors addressed, broadly applicable, able to be self-administered, and there were few commonalities in the instruments appropriate for a range of age and cultural groups, and selected to measure health behaviors. The initial P4H that have suffi cient validity and reliability. Also, mea- projects proposed the following behavioral measures. sures should be congruent with national public health goals and sensitive to intervention effects, because the Physical Activity goal is to produce improvements in individual health as Physical activity logs, recalls, weekly assessments, and well as affect population health. screening questions were all used. Referenced question- Measures for primary care research need to be brief naires or sources included Physical Activity Scale for the 22,23 24 because of the lack of time or supervision to complete Elderly (PASE), PACE project assessment, Summary 25 3 lengthy instruments. Additionally, because primary of Diabetes Self-Care Activities, Healthy People 2010, 18,19 26 care is comprehensive —concerned with improving National Health Interview Survey (NHIS), Behavioral a variety of risk factors and conditions—focusing on Risk Factor Surveillance System—BRFSS, and Guide- 1 or 2 areas should not result in worse care in other lines for Adolescent Prevention Services (GAPS). areas. Unfortunately, almost all “gold standard” assess- ment procedures have focused on 1 health behavior Risky Drinking in isolation and attain part of their precision by being Number of alcoholic drinks per week, quantity of relatively lengthy, burdensome, or expensive. drinks, social aspects of drinking, and alcohol exposure In this article, we focus on assessment of 4 health and risk (adolescents only) assessments were all used. behaviors that collectively account for the majority Referenced questionnaires or sources included the Five of illness and mortality in the United States: cigarette Shot questionnaire, Alcohol Use Disorders Identifi ca- 30,31 26 27 28 smoking, eating patterns, physical inactivity, and risky tion Test (AUDIT), NHIS, BRFSS, and GAPS. 8,20 drinking. This work was conducted as part of Pre- scription for Health (P4H), a program of the Robert Cigarette Smoking Wood Johnson Foundation and the Agency for Health- Questions included whether the patient ever smoked, care Research and Quality. In its fi rst round, P4H was a current smoker, number cigarettes smoked and funded 17 practice-based research networks to fi eld quit attempts. Referenced questionnaires or sources test evidence-based strategies to improve the delivery, included Summary of Diabetes Self-Care Activities, 26 27 28 feasibility, and reach of behavior change counseling for NHIS, BRFSS, and GAPS. these 4 health behaviors in primary care practices. A major goal of the second round is to assess the extent Eating Patterns to which interventions are effective in attaining patient Food diaries, logs, 24-hour dietary recall, and nutrition behavior change. A common set of brief measures is intake questionnaires were used. Referenced sources necessary to compare outcomes across settings, inter- included the Summary of Diabetes Self-Care Activities, 32 33 ventions, and populations. Block Brief 2000 Questionnaire, “Rate Your Plate,” 26 27 3 28 Our purposes in this article are to (1) discuss issues NHIS, BRFSS, Healthy People 2010, and GAPS. in selecting measures of physical activity, risky drink- The length of the proposed assessment instruments ing, cigarette smoking, and eating patterns that are was surprising. Subsequent conversations with P4H appropriate for adult or pediatric primary care research; grantees suggested that it was diffi cult to implement (2) identify specifi c, practical behavioral assessments these measurement tools in practice. Although many ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 74 PRACTICAL HEALTH BEHAVIOR MEASURES (NIH) Web sites, using the Table 1. Criteria Used to Select Health Behavior Change Assessments search terms “nutrition,” “diet,” “eating,” “smoking,” “cigarette Criteria Rationale/Description smoking,” “alcohol consumption,” Primary “risky drinking,” “exercise,” and 1. Practicality A composite criterion including length; ease of administration, scor- ing, and interpretation; appropriateness to a wide range of set- “physical activity” with “self- tings and populations; and in the public domain, to reduce costs report,” “measurement,” “primary 2. Relationship to clinical Assessment of success toward achieving quantifi able public health care,” and “assessment.” In partic- and public health goals goals linked to health behavior change objectives. For example, preference was given to results that could be linked to achieve- ular, reviews from 2 major NIH- ment of Healthy People 2010 goals funded health behavior change 3. Validity Agreement with reference standard measures. We gave moderate consortia, the Behavior Change emphasis to convergent and divergent validity (correlations with other established measures of the same behavior and lack of cor- 34 Consortium and the Health relation with other constructs) Maintenance Consortium (http:// 4. Sensitivity to change Ability of a measure to detect improvements resulting from effective interventions was given the largest weight. To achieve sensitivity hmcrc.srph.tamu.edu/default. to change, an instrument must demonstrate other characteristics, aspx), provided excellent infor- such as test-retest reliability. The primary purpose of including behavioral measures was to evaluate whether signifi cant improve- mation on potential instruments. ments are produced An initial subset of measures 5. Relation to behavioral Degree to which instruments were behaviorally based and thus 21 was identifi ed by the fi rst 3 interventions directly related to P4H interventions (eg, it is less valuable to have a recall that produces detailed dietary nutrients, than mea- authors, with exclusion of mea- sures that assess changes in eating behavior) sures that were longer than 10 Secondary items or that were not appropri- 6. Cultural and literacy Extent to which measures were appropriate for and had been trans- ate for self-administration. Can- issues lated into and validated in other languages, especially Spanish didate measures were initially 7. Norms and breadth Having data on national or regional norms. Data from instruments of use with well-established norms allow comparison of interventions reviewed to establish their suit- and populations to larger reference groups ability, and tables were devel- 8. Age appropriateness Appropriate across the full age range and functional levels. Given special attention for children and adolescents oped that outlined information 9. Reliability Given the brevity required, we did not place strong emphasis on on the various selection criteria. internal consistency (α coeffi cients), but gave considerable weight Summary information and to test-retest reliability tables were presented to a con- 10. Usability for feedback Scales that can be scored immediately or can identify areas for and interventions improvement to help with patient goal setting sensus panel of P4H program representatives and invited of the proposed instruments had demonstrated validity experts. The panel included 12 experts in the areas of for screening, their ability to detect changes in health primary care, practice-based research, and quantitative behaviors is uncertain. and qualitative analyses. The panel agreed that priority for further recommendations should be based on assess- Procedures and Criteria Used ments demonstrated to have (1) sensitivity to change in Making Recommendations or intervention effects; (2) broad applicability, espe- To identify practical measures for assessing and track- cially to minority populations or validated in Spanish; ing health behaviors, we needed to specify selection (3) normative data; and (4) usefulness for intervention criteria. Greatest weight was given to 5 primary crite- planning. Final revised recommendations were based on ria: practicality, clinical and public health relevance, discussion by this consensus panel. validity, sensitivity to change, and relation to behav- The following key issues emerged from the panel ioral intervention. The 5 secondary criteria listed at the discussion: (1) the number of items needed to be even bottom of Table 1 were also considered. Because child briefer than originally proposed if the expectation was and adolescent risk behaviors differ in frequency and that all practices would assess multiple behaviors; (2) nature from those of adults, separate assessments were the wording of several items did not apply to primary generally necessary. care; (3) response formats of several items needed to be Candidate measures were identifi ed from several changed to fi t time frames relevant to practice-based sources, including instruments implemented in the fi rst research; (4) a recommended list of brief items should phase of P4H, review of published literature, sugges- be provided, as well as a secondary recommendation tions of professional contacts (because many instruments where warranted, of slightly longer instruments if they are under development) and consensus statements and had superior qualities; and (5) sensitivity to change was guidelines for recommended assessments. Searches were very important, and several widely used screening mea- conducted on PubMed and National Institute of Health sures did not meet this criterion. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 75 PRACTICAL HEALTH BEHAVIOR MEASURES Figure 1. Adult Measures of Behavior Change Physical activity (RAPA) * Scoring – see Web site: http://hmcrc.srph.tamu.edu/Measurement%20folder/PA/RAPA%20fnl%201004.doc Physical activities are activities where you move and increase your heart rate above its resting rate, whether you do them for pleasure, work, or transportation. The following questions ask about the amount and intensity of physical activity you usually do. The intensity of the activity is related to the amount of energy you use doing these activities. Examples of physical activity intensity levels: Light activities Your heart beats slightly faster than normal You can talk and sing Light exercise Light stretching Light vacuuming or yard work Moderate activities Your heart beats faster than normal You can talk but not sing Brisk walking Aerobics class Strength training Swim gently Vigorous activities Your heart rate increases a lot You can’t talk, or your talking is broken up by large breaths Aerobics classes Jogging, Running, Singles tennis, or Power Walking Racquetball, Pickle ball How physically active are you? (answer Yes or No to each) Yes No I rarely or never do any physical activities. I do some light and/or moderate physical activities, but not every week. I do some light physical activity every week. I do moderate physical activity every week but less than 5 days per week or less than 30 minutes on those days. I do vigorous physical activities every week, but less than 3 days per week or less than 20 minutes on those days. I do 30 minutes or more per day of moderate physical activities 5 or more days per week. I do 20 minutes or more per day of vigorous physical activities 3 or more days per week. I do activities to increase muscle strength, such as lifting weights or calisthenics, once a week or more. I do activities to improve fl exibility, such as stretching or yoga, once a week or more. *Adapted with permission from “How Physically Active Are You?” from the University of Washington Health Promotion Research Center, Seattle, Wash. Figure 1 continues ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 76 PRACTICAL HEALTH BEHAVIOR MEASURES Risky drinking (BRFSS 2003 ) Scoring – Number drinks. Binge drinking = 5 or more drinks for men; 4 or more for women A drink of alcohol is defi ned as 1 can or bottle of beer, 1 glass of wine, 1 can or bottle of wine cooler, 1 cocktail, or 1 shot of liquor. During the past 30 days, how many days per week or per month did you have at least 1 drink of any alcoholic beverage? [if none, STOP] _______________ On the days when you drank, about how many drinks did you drink on average? ______________ Considering all types of alcoholic beverages, how many times during the past 30 days did you have 5 or more drinks on an occasion? (for women the threshold for binge drinking is 4 drinks). _________________ Cigarette smoking Scoring – any current smoking, especially daily smoking as trigger for intervention Have you smoked at least 100 cigarettes in your entire life? [if no, STOP] (BRFSS 2003) _______________ Have you smoked at least part of a cigarette in the last 7 days? [if no, STOP] (SRNT) _______________ During a typical 7-day period how many cigarettes did you smoke? (BCC) _______________ Eating patterns – (Starting the Conversation – Diet) Scoring — Contact Alice Ammerman at Alice_Ammerman@unc.edu (personal communication, May 15, 2004) How many times a week do you eat fast food meals or snacks? ___ 1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many servings of fruit or vegetables do you eat each day? ___ 1 serving ___ 2 servings ___3 or more servings ___ None ___ Don’t know/not sure How many regular sodas or glasses of sweet tea do you drink each day? (one glass is an 8 oz serving) ___<1 glass ___ 1 glass ___ 2 glasses ___ 3 glasses ____4 or more ___ None ___Don’t know/not sure How many times a week do you eat beans (like pinto or black beans), chicken or fi sh? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many times a week do you eat regular snack chips or crackers (not the low-fat)? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How many times a week do you eat desserts and other sweets? ___1 time ___ 2 times ___ 3 or more times ___ None ___Don’t know/not sure How much margarine, butter or meat fat do you use to season vegetables or put on potatoes, bread, or corn? ___ Very little ___ Some ___ A lot BRFSS = Behavioral Risk Factor Surveillance System; SRNT = Society for Research on Nicotine and Tobacco; BCC = Behavior Change Consortium. RESULTS was it possible to include entire instruments. Several For each target behavior, we offer both a primary of our recommendations endorsed new instruments, recommendation, and in most cases, a secondary rec- and potential users should review these measures for 25,27,34,36-45 ommendation, which often involves an elaboration of updates before implementing. Table 3 (avail- the primary measure that might be applicable for sites able online only at http://www.annfammed.org/ wishing to devote more time or priority to that behav- cgi/content/full/3/1/73/DC1) summarizes the ior. Our primary recommendations are summarized characteristics of recommended measures. 34-36 27,37-41 in Figure 1 and Table 2 (which are available online at http://www.annfammed.org/cgi/con- Recommended Measures for Adolescents tent/full/3/1/73/DC1); secondary recommenda- and Children 38,39,42-44,45 25,27,30,31,36 tions are in Appendixes 1 and 2 Only a few measures of health behaviors in children (which are available online at http://www.annfammed. or adolescents were located that fi t most of the desired org/cgi/content/full/3/1/73/DC1). For adults, 22 items criteria, and none were identifi ed that reported sen- are recommended as a minimum assessment across sitivity to change. Some instruments with acceptable all 4 behaviors. The corresponding number for ado- validity and reliability were identifi ed, but they were lescents, ages 12 to 17 years is 16, and 12 items for too long or cumbersome to be feasible in primary younger children. Although many of the items rec- care settings. Given the paucity of available measures, ommended are from validated instruments, seldom recommended items were sometimes based on items ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 77 PRACTICAL HEALTH BEHAVIOR MEASURES from adult samples or that had face validity but no Children. Risky drinking in children younger than established measurement properties in adolescents and 12 years is infrequent, and validated measures of sus- children. ceptibility are not currently available. The following measures are recommended for assessing adolescent and child behavior. Table 3 pres- Smoking ents the actual items. Adolescents. Two items from the Smoking Uptake Continuum assessment are recommended. These Physical Activity items assess susceptibility to smoking, with the fi rst Adolescents. Three items are recommended, including item asking about experimentation with cigarettes and 2 items from the PACE+ project and 1 item from the the second asking how many days, in the last 30, that Youth Risk Behavior Surveillance System (YRBSS). an adolescent smoked. The second item can be used to The PACE+ items, which assess frequency of physi- calculate attainment of Healthy People 2010 goals. cal activity in the last 7 days and in a typical week, The secondary recommendation for adolescent have been validated in a diverse population of middle- smoking is to include 8 additional items from the school adolescents. The items demonstrated accept- Smoking Uptake Continuum (Appendix 1). This 10- able measurement properties (sensitivity 71%, specifi c- item measure categorizes youth into 7 levels of smok- ity 63%; signifi cantly correlated with accelerometer ing behavior from committed never smoker to current readings r = .40, P<.001). In addition, our panel established smoker. The instrument is validated with believed it was important to assess sedentary behav- adolescents aged 12 to 18 years and has demonstrated ior, so an item from the YRBSS was recommended to good predictive validity (<6% committed never smok- evaluate hours of television watched on an average ers progress to established smokers over a 3- to 4-year school day. All recommended items can be used to period, susceptible never smokers are 2 to 3 times more address the Healthy People 2010 goals. A secondary likely to initiate smoking over a 3- to 4-year period recommendation was to assess sedentary behaviors of than are committed never smokers). computer and video game use via an adaptation of the Children. A primary recommendation is to assess YRBSS item that assesses television viewing. Measure- children’s susceptibility to future smoking based on 4 ment characteristics and normative data are not cur- items from Jackson’s Susceptibility to Smoking scale. rently available for this item. This measure has been validated in children in grades Younger Children. Obtaining reliable and valid 3 to 5 and found to be predictive of smoking initiation self-reported physical activity in younger children is 2 years after assessment. 46,47 problematic. No feasible and reliable self-report instruments were identifi ed for children, and objective Eating Patterns measures such as activity monitors were not considered Adolescents. The literature review identifi ed a few practical. Therefore, no recommendation is made for measures that assessed nutrient intake and had been this age-group. validated in clinical settings with adolescents and chil- 49,50 dren. These instruments were considered too long, Risky Drinking however, to be feasible in primary care settings. The Adolescents. The 3 items recommended are based on panel also emphasized the importance of assessing questions from the 2003 BRFSS. While our panel rec- dietary behaviors and food groups, such as fruits and ognized that overall prevalence of alcohol use would vegetables, rather than nutrients, for use in behavioral be lower in adolescents compared to adults, these items counseling. were recommended to identify current problems with We recommend an 8-item instrument developed alcohol consumption. They are also useful for calculat- by Ammerman and colleagues (Alice Ammerman, ing Healthy People 2010 goal attainment. personal communication, May 15, 2004) that assesses As a secondary assessment of risky drinking, 6 items dietary patterns, such as frequency of fast-food intake, from the CRAFFT screening test are recommended. fat snack intake, and sweetened beverage consump- This measure has been validated in clinical settings tion. The instrument was designed to be completed with diverse populations of adolescents aged 13 to19 by adolescents on their own and together with parents 42-44 years and has demonstrated reliability and validity. for younger children. It was designed using literature An adapted version of the instrument that screens review, expert input, and feasibility testing, and mea- only for alcohol, instead of alcohol and drug use, was surement properties are currently being evaluated. believed to be most practical. Normative data are avail- A secondary recommendation is a 2-item assessment able on the longer CRAFFT instrument, but cutoffs for of fruit and vegetable intake developed by the PACE+ specifi c alcohol-related problems are not. project. These items are consistent with Healthy ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 78 PRACTICAL HEALTH BEHAVIOR MEASURES People 2010 goals and have been validated in a diverse Time frames of many of the AUDIT items prevent pre- population of middle school-aged adolescents. They cise change estimates. have demonstrated acceptable measurement properties (signifi cant correlations with 3-day food record; sen- Cigarette Smoking sitivity 81%, specifi city 47%, positive predictive value Primary measure. Three items from national health sur- 27,34,36 79%). Although the items have not been validated in veys. These items (Figure 1) are recommended to younger children, the panel believed that younger chil- assess whether respondents ever smoked, their current dren could complete these items with assistance from a smoking status, and extent of smoking. Individual items parent or guardian. are taken from national surveys, and thus there are normative data; however, with the exception of the sec- Recommended Measures in Adults ond item, these are not standard measures from major Our review indicated that there are now several practi- trials to evaluate smoking interventions. There was cal lifestyle measures being tested for use in adults, 2 of concern about wording of some of the lengthier and which are primary recommendations. We attempted to more complex items used in smoking cessation trials. select instruments that could bridge the vast diversity The recommended items form an effi cient condensed of adult populations (Figure 1). set (if persons answer “no” to initial items, they skip the remaining questions) and allow tracking on Healthy Physical Activity People 2010 goals. Primary measure. The Rapid Assessment Physical Secondary measure. Six items recommended by 35 36 Activity Scale (RAPA). This scale involves 9 yes/no the Society for Research on Nicotine and Tobacco items assessing the type and amount of physical activ- (embedded within 3 questions in Appendix 2) to mea- ity in which adults engage. Clinicians and measurement sure cessation in cessation trials. These items refl ect experts at the University of Washington developed this expert consensus assessment of abstinence parameters scale for primary care settings. In preliminary validation (eg, temporal intervals, slips). They are recommended studies it compares well with longer scales. A major for those emphasizing smoking behaviors. This approach advantage is that it enables respondents to visualize dif- recognizes that there will be relapses and also allows ferences in activity intensity. Pictures and examples can survival analyses. be made culturally and setting relevant. Other advan- tages are a quick and easy scoring sheet, and availability Eating Patterns in different languages (eg, Spanish and Vietnamese). Primary measure. Starting the Conversation (STC) Secondary measure. Seven items from the BRFSS. – Diet. (Alice Ammerman, personal communication, The main advantage of the BRFSS is that it permits May 15, 2004) This 7-item scale comes out of a pro- comparison with national norms, and Spanish trans- gram of nutrition research by Ammerman and col- lations are available. Although separating types of leagues. The advantage of this scale is its simplicity activity may be helpful, the original purpose of the and relative brevity. It also assesses food patterns vs BRFSS was for surveillance, not intervention tracking. nutrient or fat intake and provides tailored strategies Additionally, responses are often ambiguous because for behavioral counseling. The validity of the STC of respondent failure to differentiate among intensity is currently being examined in the WISEWOMAN levels. We recommend RAPA over this scale, for likeli- Study, with comparisons to longer, more traditional hood of more accurate reporting, and the use of graph- nutritional assessments. ics to clarify different levels of activity. Both scales can Secondary measure. The Summary of Diabetes be mapped to Healthy People 2010 recommendations. Self-Care Activities (SDSCA). This scale includes 4 dietary items that address the most commonly targeted Risky Drinking dietary recommendations—eg, low fat, high fi ber, and Primary measure. Three items from the BRFSS. These increased fruit and vegetable intake. The questions cor- items offer a fast, simple way to detect risky drinking relate reasonably well with much longer reference stan- and to examine changes in binge drinking. Findings dard instruments. The SDSCA has been widely used in can be compared with national norms and related to various formats (paper and computer administered) and Healthy People 2010 goals. found to be sensitive to change in intervention studies 30,31 Secondary measure. AUDIT. This 10-item with diabetes patients, but has not been validated in instrument is often used in national and international nondiabetic study samples. The questions seem appli- studies to screen for risky drinking, and it is available cable to nondiabetic adults, but we recommend the in Spanish. It is less relevant for intervention studies STC items because they lead directly to intervention and assessing change over time than it is for screening. planning. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 79 PRACTICAL HEALTH BEHAVIOR MEASURES DISCUSSION approaches, such as item response theory, is needed to Our goal was to derive a set of practical, yet valid, determine whether it is possible to further streamline items that could evaluate patient behavior change in these assessments. the 4 priority areas across the diverse P4H projects. To read or post commentaries in response to this article, see it This same set of items should be appropriate for online at http://www.annfammed.org/cgi/content/full/3/1/73. most primary care-based research and practices in the United States. Given our mandate, the failure of tradi- Key words: Health behavior; risk factors; needs assessment; review; pri- tional review procedures to identify many appropriate mary health care measures, and the frequent lack of information on key criteria (eg, sensitivity to intervention; cultural appro- Submitted September 15, 2004; submitted, revised November 29, 2004; accepted December 1, 2004. priateness), we were forced to rely to a moderate extent upon informal networks of leading researchers and Funding support: This project was supported by the Robert Wood John- organizations, and upon subjective estimates of which son Foundation in supporting the Prescription for Health project. items best fulfi lled our criteria. Consistent use of the proposed measures would References greatly improve the science of health promotion among primary care patients. It is diffi cult to compare results 1. U. S. Preventive Services Task Force. Guide to Clinical Preventive Ser- vices. 2nd ed. Baltimore, MD: Williams & Wilkins; 1996. of studies from different interventions and populations 2. Centers for Disease Control and Prevention. The Guide to Community if each study uses measures of differing or unknown Preventive Services. 2002 Available at: http://www.thecommunity- reliability, validity, and sensitivity to change—even if guide.org. results are converted to standard units such as effect 3. U. S. Department of Health and Human Services. Healthy People size. We propose the items in Figure 1 and Table 2, 2010: Understanding and Improving Health. Washington, DC: Govern- as a relatively succinct set of items for use in primary ment Printing Offi ce; 2000. 52,53 care research and practical clinical trials. These 4. Ockene IS, Hebert JR, Ockene JK, et al. Effect of physician-delivered nutrition counseling training and an offi ce-support program on satu- recommendations should also be useful to clinicians rated fat intake, weight, and serum lipid measurements in a hyper- and health plans for identifi cation of health behavior lipidemic population: Worcester Area Trial for Counseling Hyperlipid- change needs and quality improvement efforts. emia (WATCH). Arch Intern Med. 1999;159:725-731. Additional research is needed on these measures. 5. Whitlock EP, Orleans CT, Pender N, Allan J. Evaluating primary care behavioral counseling interventions: an evidence-based approach. Am Some items are very new, and more data are needed J Prev Med. 2002;22:267-284. on their continued performance. Several items recom- 6. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the mended have not been fully tested in terms of tradi- incidence of type 2 diabetes with lifestyle intervention or metformin. tional psychometric standards. There is, however, a N Engl J Med. 2002;346:393-403. compelling need for a practical set of items appropriate 7. McGinnis JM, Foege WH. Actual causes of death in the United for assessing multiple health behaviors in primary care. States. JAMA. 1993;270:2207-2212. We posit that the recommended items are a better 8. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States. JAMA. 2004;291:1238-1245. solution to this need than recommending nothing, or a much longer, and less clinically useful set of mea- 9. Institute of Medicine, Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st sures that would be much less likely to be used. These Century. Washington, DC: National Academy Press; 2001. recommendations should be periodically reevaluated 10. Luire JD, Goodman DC, Wennberg JE. Benchmarking the future gen- and revised as new information becomes available. eralist workforce. Eff Clin Pract. 2002;5:58-66. This report is limited to assessments of these 4 health 11. Shipman LA, Lurie JD, Goodman DC. The general pediatrician: behaviors and does not address other important self- projection of future workforce supply and requirements. Pediatrics. 2004;113:435-442. management goals that may be set between patients and their health care team. We restricted our focus to 12. Council on Graduate Medicine Education. Physician Workforce Policy Guidelines for the U.S. 2000-2020. Albany, New York: Center for self-report measures and did not review direct observa- Health Workforce Studies, University of Albany; 2004. tion or mechanical instrumentation methods. 13. Larson EB, Fihn SD, Kirk LM, et al. The future of general internal We recommend 3 specifi c areas of future research. medicine. Report and recommendations from the Society of General First, these proposed items should be compared directly Internal Medicine (SGIM) Task Force on the Domain of General Inter- nal Medicine. J Gen Intern Med. 2004;19:69-77. to “gold standard” assessments. Second, more studies 14. Future of Family Medicine Project Leadership Committee. The future are needed of their sensitivity to change; acceptability of family medicine: a collaborative project of the family medicine in a spectrum of family medicine, internal medicine, community. Ann Fam Med. 2004;2:S3-S32. and pediatric practices; and results under different 15. Shumaker SA, Schron EB, Ockene JK, McBee WL. The Handbook of conditions of administration, in particular computer- Health Behavior Change. 2nd ed. New York, NY: Springer Publishing automated assessment. Finally, research using modern Company; 1998. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005 80 PRACTICAL HEALTH BEHAVIOR MEASURES 16. Stange KC, Woolf SH, Gjeltema K. One minute for prevention: the 37. Prochaska JJ, Sallis JF, Long B. A physical activity screening measure power of leveraging to fulfi ll the promise of health behavior counsel- for use with adolescents in primary care. Arch Pediatr Adolesc Med. ing. Am J Prev Med. 2002;22:320-323. 2001;155:554-559. 17. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical 38. Centers for Disease Control and Prevention. National Center for Care Survey: 2001 summary. Adv Data. 2003;22:1-44. Chronic Disease Prevention and Health Promotion. Youth Risk Behav- ior Surveillance System. Available at: http://www.cdc.gov/Healthy 18. Starfi eld B. Primary care: Balancing Health Needs, Services, and Technol- Youth/yrbs/about_yrbss.htm. ogy. Cambridge: Oxford University Press; 1998. 39. Choi WS, Gilpin EA, Farkas AJ, Pierce JP. Determining the probability 19. Institute of Medicine. Primary Care: America’s Health in a New Era. of future smoking among adolescents. Addiction. 2001;2:313-323. Washington, D.C.: National Academy of Sciences; 1996. 40. North Carolina Department of Health and Human Services. Physical 20. Fine LJ, Philogene GS, Grambling R, Coups EJ, Sinha S. Prevalence activity and nutrition behaviors monitoring form. Available at: http:// of multiple behavioral risk factors in the United States: Results www.eatsmartmovemorenc.com/data/PAN%20Behaviors%20Monitori from the 2001 National Health Interview Survey. Am J Prev Med. ng%20Form.pdf 2004;27(2S):18-24. 41. Jackson C. Cognitive susceptibility to smoking and initiation 21. Robert Wood Johnson Foundation, Agency for Healthcare Research of smoking during childhood: a longitudinal study. Prev Med. and Quality. PrescriptionForHealth Web site. Available at: http://www. 1998;27:129-134. prescriptionforhealth.org. 42. Knight JR, Shrier LA, Bravender TD, Farrell M, Vander Bilt J, Shaffer 22. Washburn RA, Smith KW, Jette AM, Janney CA. The physical activity HJ. A new brief screen for adolescent substance abuse. Arch Pediatr scale for the elderly (PASE): development and evaluation. J Clin Epide- Adolesc Med. 1999;153:592-596. miol. 1993;46:153-162. 43. Knight JR, Sherritt L, Shrier La, Karris SK, Chang G. Validity of the 23. Washburn RA, McAuley E, Katula J, Mihalko SL, Boileau RA. The CRAFFT substance abuse screening test among adolescent clinic physical activity scale for the elderly (PASE): evidence for validity. J patients. Arch Pediatr Adolesc Med. 2002;156:607-614. Clin Epidemiol. 1999;52:643-651. 44. Cummings LH, Chan KK, Burns KM, Blume AW, Larimer M, Marlatt 24. Calfas KJ, Sallis JF, Zabinski MF, et al. Preliminary evaluation of GA. Validity of the CRAFFT in American-Indian and Alaska-Native multi-component program for nutrition and physical activity change adolescents: screening for drug and alcohol risk. J Stud Alcohol. in primary care: PACE+ for adults. Prev Med. 2002;34:153-161. 2003;64:727-732. 25. Toobert DJ, Hampson SE, Glasgow RE. The summary of Diabetes 45. Prochaska JJ, Sallis JF. Reliability and validity of a fruit and Self-Care Activities Measure: results from seven studies and revised vegetable screening measure for adolescents. J Adolesc Health. scale. Diabetes Care. 2000;23:943-950. 2004;34:163-165. 26. U. S. Department of Health and Human Services. National Health 46. Klesges LM, Baranowski T, Beech B, et al. Social desirability bias in Interview Survey (NHIS). 2004. Available at: http://www.cdc.gov/ self-reported dietary, physical activity and weight concerns measures nchs/nhis.htm. in 8- to 10-year old African-American girls: Results from the Girls 27. U. S. Department of Health and Human Services. Behavioral Risk Health Enrichment Multisite Studies (GEMS). Prev Med. 2004;38: Factor Surveillance System. 2004. Available at: http://www.cdc.gov/ S78-S87. brfss/index.htm. 47. Welk GJ, Corbin CB, Dale D. Measurement issues in the assessment 28. Elster A. The American Medical Association guidelines for adolescent of physical activity in children. Res Q Exerc Sport. 2000;71(2 Suppl): preventive services. Arch Pediatr Adolesc Med. 1997;151:958-959. S59-S73. 29. Seppa K, Lepisto J, Sillanaukee P. Five-shot questionnaire on heavy 48. Substance Abuse and Mental Health Services Administration. Results drinking. Alcohol Clin Exp Res. 1998;22:1788-1789. from the 2002 National Household Survey on Drug Use and Health: National Findings. Pediatrics. 2001. Rockville, MD. 30. The Alcohol Use Disorders Identifi cation Test: Guidelines for Use in Prima- ry Health Care. Geneva, Switzerland: World Health Organization, 1989. 49. Prochaska JJ, Sallis JF, Rupp J. Screening measure for assessing dietary fat intake among adolescents. Prev Med. 2001;33:699-706. 31. Saunders JB, Aasland OB, Babor TF, et al. Development of the Alcohol Use Disorders Identifi cation Test (AUDIT): WHO collaborative 50. Dennison BA, Jenkins PL, Rockwell HL. Development and valida- project on early detection of persons with harmful alcohol consump- tion of an instrument to assess child dietary fat intake. Prev Med. tion--II. Addiction. 1993; 88:791-804. 2000;31:214-224. 32. Block G, Gillespie C, Rosenbaum EH, Jenson C. A rapid food 51. Will JC, Farris RP, Sanders CG, Stockmyer CK, Finkelstein EA. Health screener to assess fat and fruit and vegetable intake. Am J Prev Med. promotion interventions for disadvantaged women: overview of the 2000;18:284-288. WISEWOMAN projects. J Womens Health. 2004;13:484-502. 33. Gans KM, Sundaram SG, McPhillips JB, Hixson ML, Linman Carlton 52. Tunis SR, Stryer DB, Clancey CM. Practical clinical trials. Increas- RA. Rate your plate: an eating pattern assessment and educational tool ing the value of clinical research for decision making in clinical and used at cholesterol screening programs. J Ntr Educ. 1993;25:29-36. health policy. JAMA. 2003;290:1624-1632. 34. Ory MG, Jordan PJ, Bazzare T. The Behavior Change Consortium: set- 53. Glasgow RE, Magid DJ, Beck A, Ritzwoller D, Estabrooks PA. Practi- ting the stage for a new century of health behavior change research. cal clinical trials for translating research to practice: design and mea- Health Educ Res. 2002;17:500-511. surement recommendations. Med Care. In press. 35. University of Washington Health Promotion Research Center. Rapid 54. Wagner EH. The role of patient care teams in chronic disease man- Assessment Physical Activity Scale (RAPA). Available at: http://hmcrc. agement. BMJ. 2000;320:569-572. srph.tau.edu. 55. Babor TF, Sciamanna CN, Pronk N. Assessing multiple risk behaviors 36. Society for Research on Nicotine and Tobacco Web site. Available in primary care: screening issues and related concepts. Am J Prev at http://hmcrc.srph.tamu.edu/Measurement%20folder/Smoking/ Med. 2004; 27(2S):42-53. Tobacco%20SNT.pdf. ✦ ✦ ✦ ANNALS OF FA MILY MEDICINE WWW.ANNFAMMED.ORG VOL. 3, NO. 1 JA NUARY/FEBRUA RY 2005

Journal

The Annals of Family MedicineUnpaywall

Published: Jan 1, 2005

There are no references for this article.