Abstract
Introduction
Tobacco use is a primary cause of morbidity and mortality in the United States.1,2 Multiple studies have shown that patients who smoke cigarettes are 1.6 to 3.4 times more likely to quit when their physician advises cessation.3-6 However, observed rates of tobacco cessation counseling in primary care clinics are only 20% to 30%. 7 Using a preencounter reminder may encourage more counseling and thereby increase rates of cessation. Most studies use retrospective chart review to determine if counseling had occurred during an office visit. However, even if documented, patients often do not recall specific details related to a smoking cessation plan. 8 A provider prompt could structure communication between the patient and provider in such a way that patients are more likely to remember important messages. Measuring a patient’s perception of being counseled on cessation is likely a more important outcome than whether the chart record indicates the counseling was done. Our hypothesis was that a preencounter physical reminder for the health care provider would increase a patient’s perception of smoking cessation counseling.
Methods
Participants were patients at a single-site university-affiliated family medicine residency program. The clinic provides care to more than 4800 patients; most are adults (67%) and are white (54%), black (20%), or Hispanic (21%). Most patients (68%) have public insurance, 18% have private insurance, and 13% are self-pay. Twenty-four percent of adult clinic patients smoke cigarettes. The study was approved by the BayCare Health System Institutional Review Board prior to participant enrollment.
Patients were eligible if they were ≥18 years old and were literate in English or Spanish. Enrollment occurred immediately after the office visit once informed consent was obtained. This was a quasi-experimental intervention study with a pre–post test design. A convenience sample of participants was obtained both pre- and postintervention. The length of sampling was 5 weeks pre- and 10 weeks postimplementation in order to collect the target number of survey responses.
Participants completed a survey modified from the validated Smoking Cessation by General Practitioners (SmoCess-GP) instrument. 9 The primary modification was addition of an initial question related to tobacco use in the previous 2 weeks. This allowed only data from patients who smoke to be analyzed. The remainder of the survey included the following statements: “My practitioner asked me how many cigarettes I smoke per day,” “My practitioner cautioned me about the negative consequences of smoking,” “My practitioner strongly advised me to quit smoking,” “My practitioner recommended a smoking cessation course,” “My practitioner gave me behavioral advice about quitting,” and “My practitioner prescribed or advised me to consider medication to help me quit smoking.” These items were answered yes or no by the participants.
The intervention employed the quitline card provided by the American Academy of Family Physicians displaying the 1-800-QUIT-NOW telephone number and a link to the Smokefree.gov Web site developed by the National Cancer Institute. During the intervention period, a medical assistant attached the quitline card to the back of the clinical encounter form if a patient’s intake questionnaire indicated current tobacco use.
Individual surveys were collected and reviewed for eligibility and completeness. Surveys were excluded from the analysis for the following reasons: participants had not smoked within 2 weeks, there was no encounter number on the survey, the encounter form did not document that the intervention was applied, or the survey was a duplicate. Unanswered survey items were counted as missing values and treated as no responses. Demographic information collected included age, gender, and race/ethnicity.
Categorical data analysis used chi-square and Fisher’s exact tests. The 2 independent groups were compared to assess differences in the frequency of demographic characteristics and for differences in the proportion of yes responses for individual survey questions. For individual elements, numerical values were assigned (yes = 1, no = 0) to generate summary scores reflecting the strength of patient perception that counseling took place during the visit. The difference between groups in mean summary score was compared using the Student’s t test. A change of 1 point in the mean summary score was considered to be clinically significant. Using this effect size and an estimate of baseline mean (± standard deviation) on which to improve of 2.3 (±1.58), it was calculated that 40 participants would be needed in each group to have 80% power. Analyses were conducted using SAS version 9.3. A P value ≤.05 was considered statistically significant.
Results
Of 99 surveys collected preintervention, 59 were excluded from analysis because participants did not indicate current tobacco use. Postintervention, 88 surveys were collected, with 4 excluded for duplication and 38 for being completed by nonsmoking patients. The final evaluable sample included 40 patients preintervention and 46 patients postintervention. Only 5 survey questions were counted as missing due to nonresponse. There were no significant differences between groups with regard to age, gender, or race/ethnicity (Table 1). The majority of participants were female (59%) and white (78%).
Frequency of Participant Demographic Characteristics Before Clinic Intervention (n = 40) and After Implementation of a Counseling Prompt (n = 46).
Chi-square test.
Fisher’s exact test.
The proportion of survey questions with yes responses is shown in Table 2. For most counseling elements, a high proportion of patients indicated that these items were discussed by the provider at baseline, without significant improvement postintervention. In contrast, only 40% of respondents at baseline indicated that the provider recommended a smoking cessation course. Those who perceived this aspect of counseling took place postintervention increased to 83% (P < .05). Furthermore, the mean number of elements perceived to have been addressed during the office visit increased from 4.1 to 5.1 (P < .05).
Frequency of Participant Survey Responses and Mean Survey Score Before (n = 40) and After (n = 46) a Clinic-Based Provider Counseling Prompt.
Fisher’s exact test.
Chi-square test.
T test.
Discussion
In this prospective evaluation of a practice-level intervention, there was significant improvement in patient perception that a smoking cessation course was recommended by the provider. Also, the mean number of counseling points perceived to have occurred increased significantly after a clinical process change. This positive outcome was achieved with a simple visual cue at no additional cost. As a quality improvement study, this intervention was examined as it was used in routine clinical practice. This type of outcomes research is important to determine benefits of interventions in real-world situations. 10 In addition, our results are strengthened by using a control group in the preintervention phase that was similar in characteristics to the group in the postintervention phase. 11 This increases the likelihood that the observed improvements are the result of the actual intervention and not because of inherit differences between the groups.
Previous studies of smoking cessation counseling have focused on the provision of counseling by the health care provider. It has been shown that provider education and training increases the level of comfort and familiarity with clinical practice guidelines for tobacco abuse. 12 The current study is unique in that the outcomes are patient oriented. This adds to existing literature demonstrating that the strength of physician-patient communication impacts outcomes. 6
The conceptual model of enablers, such as the reminder prompt used in this study, suggests that more research is needed on ways these interventions can be implemented and resulting effects on patient outcomes. 13 Our study shows that office staff can effectively assist physicians by providing simple prompts for patient counseling. The physical presence of this prompt may make providers more likely to perform brief smoking cessation counseling even during a busy office visit or one seemingly unrelated to tobacco use. It is also a source of additional information and aid once the patient leaves the office. Furthermore, patient recall of this counseling is improved when something tangible, such as a quitline card, is provided.
Limitations of this study should be recognized. Selection bias was possible during the enrollment procedure. A convenience sampling method was used to avoid disruptive changes in clinic operations. Although this method can be affected by changing patient characteristics over time, tobacco abuse is unlikely to be associated with such short-term variation; therefore, the sample should be representative of our overall clinic population. Also, patient recall of counseling that took place during the visit could have been inaccurate. Although an 18% false-positive recall has been reported in previous studies, we surveyed patients immediately after the office visit thereby limiting the risk of overreporting that is associated with delayed assessment. 14 Last, the high baseline rate of perceived counseling (88%) may have lessened the anticipated effect size for individual counseling elements and thus made it difficult to find significant differences in these outcomes. Nonetheless, significant findings in aggregate survey scores and measurements in regard to individual elements set a new baseline on which subsequent studies may be designed.
Conclusion
This quality improvement study shows that a visual cue improves overall patient perception of provider counseling. The quitline service reference card is an effective and inexpensive means of prompting a physician to discuss smoking cessation and can be easily integrated into practice. Specifically, it improved the unique patient perception that a smoking cessation course was recommended. Our results show that enhanced communication can change patient perceptions regarding preventive services. We believe this is an important patient-oriented outcome measure with the potential to improve community health through education.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.