- Email: [email protected]
Strategies for distributing cancer screening decision aids in primary care
Patient Education and Counseling 78 (2010) 166–168
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Medical Decision Making
Strategies for distributing cancer screening decision aids in primary care§ Charles Brackett a,*, Stephen Kearing b, Nan Cochran c, Anna N.A. Tosteson b, W. Blair Brooks a a b c
Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH, USA White River Junction, VAH, VT, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 31 October 2008 Received in revised form 19 June 2009 Accepted 29 June 2009
Objective: Decision aids (DAs) have been shown to facilitate shared decision making about cancer screening. However, little data exist on optimal strategies for dissemination. Our objective was to compare different decision aid distribution models. Methods: Eligible patients received video decision aids for prostate cancer (PSA) or colon cancer screening (CRC) through 4 distribution methods. Outcome measures included DA loans (N), % of eligible patients receiving DA, and patient and provider satisfaction. Results: Automatically mailing DAs to all age/gender appropriate patients led to near universal receipt by screening-eligible patients, but also led to ineligible patients receiving DAs. Three different elective (non-automatic) strategies led to low rates of receipt. Clinician satisfaction was higher when patients viewed the DA before the visit, and this model facilitated implementation of the screening choice. Regardless of timing or distribution method, patient satisfaction was high. Conclusions: An automatic DA distribution method is more effective than relying on individual initiative. Enabling patients to view the DA before the visit is preferred. Practice implications: Systematically offering DAs to all eligible patients before their appointments is the ideal strategy, but may be challenging to implement. ß 2009 Elsevier Ireland Ltd. All rights reserved.
Keywords: Shared decision making Decision aids Patient education and counseling Cancer screening
1. Introduction Whether or not to get screened for prostate cancer with the PSA blood test and which method to use for screening for colo-rectal cancer (CRC) are examples of preference-sensitive decisions. For such decisions there is no clear ‘‘right’’ choice due to trade offs between different options and incomplete scientific evidence regarding outcomes [1–5]. Preference-sensitive decisions are best made through shared decision making (SDM), a process in which an informed patient and their doctor reach a decision based on the patient’s specific characteristics and values [6]. Relevant professional organizations recommend SDM for these cancer screening decisions [1–5], but it is challenging to accomplish this process in the context of an office visit. Barriers to integrating SDM into routine primary care include lack of physician time and reimbursement for the lengthy discussions required to adequately
§ This research was performed at Dartmouth-Hitchcock Medical Center in Lebanon, NH and White River Junction VA Hospital in White River Junction, VT between June 2006 and May 2008. * Corresponding author at: Section of General Internal Medicine, Dartmouth Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756, USA. Tel.: +1 603 650 2921. E-mail address: [email protected] (C. Brackett).
0738-3991/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2009.06.013
address complex screening decisions and patient difficulties with comprehending the relevant information [7]. Patient decision aids (DAs) can facilitate SDM by providing balanced, standardized, evidence-based information and helping clarify values and preferences [8]. In addition to increasing knowledge and lowering decisional conflict, DAs have been shown to increase patient participation in decision making [9]. DAs also have the potential to save physicians time as they can be used outside of the clinic visit, thereby reducing the estimated 7.4 h/day of primary care clinicians’ time required to fully satisfy the U.S. Preventive Services Task Force’s recommendations [10]. There are logistical challenges to using DAs in primary care. These include limited appointment times with multiple competing agendas, identifying patients appropriate for the DA, and difficulty ‘‘closing the loop’’ (arriving at and implementing a final screening choice after viewing a DA). Little data exist on optimal strategies for dissemination. Our objective was to compare the impact of different cancer screening DA distribution models on rate of receipt by eligible patients and on clinician and patient satisfaction. 2. Methods The study took place at 2 rural academic medical centers: Dartmouth-Hitchcock Medical Center (DHMC) and the White River
C. Brackett et al. / Patient Education and Counseling 78 (2010) 166–168
167
Table 1 Methods for distributing cancer screening decision aids. Distribution method
DA
Timing
Loans, N
Eligible pts receiving DAa (%)
(1) (2) (3) (4)
PSA CRC PSA PSA CRC
Pre-visit Pre-visit Post-visit Post-visit
1625 84 724 52 33
100 6 8 2 2
a
Automatic: DA mailed to all age eligible men Elective: Letter offers DA to all age eligible, mailed upon patient request Elective: Medical assistant offers DA to eligible, patient takes home Elective: Physician offers DA to eligible, patient takes home
Eligible patient denominators estimated from patient visit counts.
Junction Veterans Administration Hospital (WRJVA). Eligible patients were loaned video DAs for prostate cancer (PSA) or colon cancer screening (CRC). These DAs, developed by the Foundation for Informed Medical Decision Making, provide information about cancer screening options, explain the importance of taking personal values into consideration and provide balanced testimonials from patients who made different cancer screening decisions. We used 4 DA distribution methods, reflecting scheduling practices at the study sites (the existence of preventive medicine visits at DHMC but not WRJVA) and patient eligibility for the different DAs (80% for PSA, 20% for CRC, based on pilot data). All patients watched the DA at home, and completed a paper-based questionnaire. Method 1: Automatic/pre-visit. The PSA DA was mailed to all men aged 50–75 before their scheduled preventive medicine visit at DHMC. Chart review for our pilot study indicated approximately 20% of men in this age group would be ineligible for the DA based on our exclusion criteria of previous diagnosis of prostate cancer, previous prostate biopsy, testosterone therapy, or family history. For this study, we decided it was more cost effective to omit chart review and ask patients to selfselect for exclusion criteria listed in a cover letter. Method 2: Elective/pre-visit. A letter was mailed to all men and women aged 50–75 prior to a scheduled preventive medicine visit at DHMC, offering the CRC DA to those eligible. Patients self-selected based on exclusion criteria (colonoscopy within 10 years, past polyp or CRC diagnosis, family history) and requested the DA, which was mailed before the visit. Method 3: Elective/post-visit. Men aged 50–75 were screened for eligibility for the PSA DA by medical assistants after any primary care visit at the WRJVA. Eligible patients were offered the DA while checking out from the clinic. Method 4: Elective/post-visit. Eligible patients were prescribed the PSA or CRC DA by their clinician at their visit. Patients picked up the DA from a lending library in the hospital after the visit. 2.1. Measures The primary outcome measures were the number of DA loans (N), % of eligible patients receiving the DA, % of those receiving the DA who return the questionnaire (a proxy for viewing the DA), and patient and provider satisfaction as assessed by written questionnaire. The number of eligible patients was determined from scheduling data and estimates of the percent eligible. 3. Results From June 2006 to May 2008, 2518 PSA and CRC decision aids were distributed to patients (Table 1). The automatic method of mailing PSA DAs to all age and gender appropriate patients led to near universal receipt among eligible patients, but also led to an estimated 20% of DAs being sent to ineligible patients. The
elective methods, which were dependent on the initiative of patients or clinicians, led to distribution rates of fewer than 10% of eligible patients. Questionnaire return rates were similar between the 4 distribution models (24–36%), suggesting that video viewing rates were also similar. Actual viewing rates exceeded questionnaire return rates, as 100% of a small sample of patients who did not return the questionnaire (n = 28) indicated that they had viewed the DA but preferred not to participate in research. In this study, pre-visit models led to higher clinician satisfaction with DAs (in a written survey, 68% of 18 DHMC clinicians exposed to pre-visit model vs. 19% of 16 WRJVA clinicians exposed to postvisit model were satisfied). Physicians stated that when their patient watched the PSA DA before the visit, it saved time during the visit and changed the conversation from the communication of facts to a discussion of values and preferences. The pre-visit model also facilitated implementation of the decision choice at the time of the visit by allowing the patient to arrive at their appointment prepared to make a cancer screening decision. Patients receiving a DA after their appointment had to communicate their screening choice with their provider at a second appointment or through other follow up channels. Regardless of timing or method of distribution, nearly all patients indicated DAs were helpful in their decision making (86– 96%) and would recommend them to other patients (94–100%). 4. Discussion and conclusion 4.1. Discussion When encouraging adoption of a desired behavior, in this case the use of DAs, inertia is a barrier to overcome. It has been shown that default (automatic) options are considerably more likely to be chosen than options requiring some initiative, however minimal [11]. Our results show that automatic delivery of DAs led to a high rate of reaching eligible patients, while distribution strategies that required the initiative of the patient or their physician led to low rates. While method 3 may seem automatic, it relied on the initiative of medical assistants who were distracted by other duties and were only able to screen for eligibility and distribute DAs to a minority of patients. A system that determines eligibility at the time of the visit may be simpler to administer, but does not allow the advantages of the patient viewing the DA before the visit. When the patient arrives at the visit having viewed the DA, the conversation with the clinician is richer, decision making is more efficient and effective, and the decision choice can be implemented at that time. Identifying all patients eligible for cancer screening and delivering DAs to them before a visit presents logistical challenges and requires more resources. In a model system, data from an electronic medical record and computerized scheduling data could be merged to efficiently identify eligible patients before the visit, and avoid the problem of distributing DAs to patients meeting exclusion criteria.
168
C. Brackett et al. / Patient Education and Counseling 78 (2010) 166–168
4.2. Conclusion
Acknowledgements
An automatic DA distribution method is more effective than relying on individual initiative. Enabling patients to view the DA before the visit is preferred.
We are grateful to Martha Coutermarsh, RN, Katrina Whyman, Inger Imset, Ephrem Micaiah, MPH and the primary care service of the WRJVA for their work on this project.
4.3. Practice implications
References
Cancer screening DAs should be systematically offered to all eligible patients before the visit with their clinician. However, this presents logistical challenges in getting DAs to the right patients at the right time. Role of funding The Foundation for Informed Medical Decision Making provided funding for this research, and had no involvement in the study design, collection, analysis and interpretation of data, writing of the report, or decision to submit for publication. FIMDM developed the decision aids used in the study. Conflict of interest The authors have no conflicts of interest, other than receiving funding as above.
[1] Screening for prostate cancer: US Preventive Services Task Force Recommendation statement. Ann Intern Med 2008;149:185–91. [2] American College of Physician. Screening for prostate cancer. Ann Intern Med 1997;126:480–4. [3] American Academy of Family Physicians. Appendix F, summary of policy recommendations for periodic health exam. In: AAFP reference manual. Kansas City, MO: AAFP; 1998. p. 62. [4] American Urological Association. Prostate specific antigen best practice policy. Oncology 2000;14:267–80. [5] US Preventive Services Task Force. Screening for colorectal cancer: recommendation and rationale. Ann Intern Med 2002;137:129–31. [6] Barry MJ, Mulley AG, Fowler FJ, Wennberg JW. Watchful waiting vs. immediate transurethral resection for symptomatic prostatism. JAMA 1988;259:3010–7. [7] Barry M. Health decision aids to facilitate shared decision making in office practice. Ann Intern Med 2002;136:127–35. [8] O’Connor A, Legare F, Stacey D. Risk communication in practice: the contribution of decision aids. BMJ 2003;327:736–40. [9] O’Connor A, Rostom A, Fiset V, Tetroe J, Entwistle V, Llewellyn-Thomas H, et al. Decision aids for patients facing health treatment or screening decisions: systematic review. BMJ 1999;319:731–4. [10] Yarnall K, Pollack K, Ostbe T, Krause K. Primary care: is there enough time for prevention? Am J Public Health 2003;93:635–41. [11] Thaler RH, Sunstein CR. Nudge: improving decisions about health, wealth, and happiness. Caravan Books; 2008.
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Medical Decision Making
Strategies for distributing cancer screening decision aids in primary care§ Charles Brackett a,*, Stephen Kearing b, Nan Cochran c, Anna N.A. Tosteson b, W. Blair Brooks a a b c
Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH, USA White River Junction, VAH, VT, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 31 October 2008 Received in revised form 19 June 2009 Accepted 29 June 2009
Objective: Decision aids (DAs) have been shown to facilitate shared decision making about cancer screening. However, little data exist on optimal strategies for dissemination. Our objective was to compare different decision aid distribution models. Methods: Eligible patients received video decision aids for prostate cancer (PSA) or colon cancer screening (CRC) through 4 distribution methods. Outcome measures included DA loans (N), % of eligible patients receiving DA, and patient and provider satisfaction. Results: Automatically mailing DAs to all age/gender appropriate patients led to near universal receipt by screening-eligible patients, but also led to ineligible patients receiving DAs. Three different elective (non-automatic) strategies led to low rates of receipt. Clinician satisfaction was higher when patients viewed the DA before the visit, and this model facilitated implementation of the screening choice. Regardless of timing or distribution method, patient satisfaction was high. Conclusions: An automatic DA distribution method is more effective than relying on individual initiative. Enabling patients to view the DA before the visit is preferred. Practice implications: Systematically offering DAs to all eligible patients before their appointments is the ideal strategy, but may be challenging to implement. ß 2009 Elsevier Ireland Ltd. All rights reserved.
Keywords: Shared decision making Decision aids Patient education and counseling Cancer screening
1. Introduction Whether or not to get screened for prostate cancer with the PSA blood test and which method to use for screening for colo-rectal cancer (CRC) are examples of preference-sensitive decisions. For such decisions there is no clear ‘‘right’’ choice due to trade offs between different options and incomplete scientific evidence regarding outcomes [1–5]. Preference-sensitive decisions are best made through shared decision making (SDM), a process in which an informed patient and their doctor reach a decision based on the patient’s specific characteristics and values [6]. Relevant professional organizations recommend SDM for these cancer screening decisions [1–5], but it is challenging to accomplish this process in the context of an office visit. Barriers to integrating SDM into routine primary care include lack of physician time and reimbursement for the lengthy discussions required to adequately
§ This research was performed at Dartmouth-Hitchcock Medical Center in Lebanon, NH and White River Junction VA Hospital in White River Junction, VT between June 2006 and May 2008. * Corresponding author at: Section of General Internal Medicine, Dartmouth Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756, USA. Tel.: +1 603 650 2921. E-mail address: [email protected] (C. Brackett).
0738-3991/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2009.06.013
address complex screening decisions and patient difficulties with comprehending the relevant information [7]. Patient decision aids (DAs) can facilitate SDM by providing balanced, standardized, evidence-based information and helping clarify values and preferences [8]. In addition to increasing knowledge and lowering decisional conflict, DAs have been shown to increase patient participation in decision making [9]. DAs also have the potential to save physicians time as they can be used outside of the clinic visit, thereby reducing the estimated 7.4 h/day of primary care clinicians’ time required to fully satisfy the U.S. Preventive Services Task Force’s recommendations [10]. There are logistical challenges to using DAs in primary care. These include limited appointment times with multiple competing agendas, identifying patients appropriate for the DA, and difficulty ‘‘closing the loop’’ (arriving at and implementing a final screening choice after viewing a DA). Little data exist on optimal strategies for dissemination. Our objective was to compare the impact of different cancer screening DA distribution models on rate of receipt by eligible patients and on clinician and patient satisfaction. 2. Methods The study took place at 2 rural academic medical centers: Dartmouth-Hitchcock Medical Center (DHMC) and the White River
C. Brackett et al. / Patient Education and Counseling 78 (2010) 166–168
167
Table 1 Methods for distributing cancer screening decision aids. Distribution method
DA
Timing
Loans, N
Eligible pts receiving DAa (%)
(1) (2) (3) (4)
PSA CRC PSA PSA CRC
Pre-visit Pre-visit Post-visit Post-visit
1625 84 724 52 33
100 6 8 2 2
a
Automatic: DA mailed to all age eligible men Elective: Letter offers DA to all age eligible, mailed upon patient request Elective: Medical assistant offers DA to eligible, patient takes home Elective: Physician offers DA to eligible, patient takes home
Eligible patient denominators estimated from patient visit counts.
Junction Veterans Administration Hospital (WRJVA). Eligible patients were loaned video DAs for prostate cancer (PSA) or colon cancer screening (CRC). These DAs, developed by the Foundation for Informed Medical Decision Making, provide information about cancer screening options, explain the importance of taking personal values into consideration and provide balanced testimonials from patients who made different cancer screening decisions. We used 4 DA distribution methods, reflecting scheduling practices at the study sites (the existence of preventive medicine visits at DHMC but not WRJVA) and patient eligibility for the different DAs (80% for PSA, 20% for CRC, based on pilot data). All patients watched the DA at home, and completed a paper-based questionnaire. Method 1: Automatic/pre-visit. The PSA DA was mailed to all men aged 50–75 before their scheduled preventive medicine visit at DHMC. Chart review for our pilot study indicated approximately 20% of men in this age group would be ineligible for the DA based on our exclusion criteria of previous diagnosis of prostate cancer, previous prostate biopsy, testosterone therapy, or family history. For this study, we decided it was more cost effective to omit chart review and ask patients to selfselect for exclusion criteria listed in a cover letter. Method 2: Elective/pre-visit. A letter was mailed to all men and women aged 50–75 prior to a scheduled preventive medicine visit at DHMC, offering the CRC DA to those eligible. Patients self-selected based on exclusion criteria (colonoscopy within 10 years, past polyp or CRC diagnosis, family history) and requested the DA, which was mailed before the visit. Method 3: Elective/post-visit. Men aged 50–75 were screened for eligibility for the PSA DA by medical assistants after any primary care visit at the WRJVA. Eligible patients were offered the DA while checking out from the clinic. Method 4: Elective/post-visit. Eligible patients were prescribed the PSA or CRC DA by their clinician at their visit. Patients picked up the DA from a lending library in the hospital after the visit. 2.1. Measures The primary outcome measures were the number of DA loans (N), % of eligible patients receiving the DA, % of those receiving the DA who return the questionnaire (a proxy for viewing the DA), and patient and provider satisfaction as assessed by written questionnaire. The number of eligible patients was determined from scheduling data and estimates of the percent eligible. 3. Results From June 2006 to May 2008, 2518 PSA and CRC decision aids were distributed to patients (Table 1). The automatic method of mailing PSA DAs to all age and gender appropriate patients led to near universal receipt among eligible patients, but also led to an estimated 20% of DAs being sent to ineligible patients. The
elective methods, which were dependent on the initiative of patients or clinicians, led to distribution rates of fewer than 10% of eligible patients. Questionnaire return rates were similar between the 4 distribution models (24–36%), suggesting that video viewing rates were also similar. Actual viewing rates exceeded questionnaire return rates, as 100% of a small sample of patients who did not return the questionnaire (n = 28) indicated that they had viewed the DA but preferred not to participate in research. In this study, pre-visit models led to higher clinician satisfaction with DAs (in a written survey, 68% of 18 DHMC clinicians exposed to pre-visit model vs. 19% of 16 WRJVA clinicians exposed to postvisit model were satisfied). Physicians stated that when their patient watched the PSA DA before the visit, it saved time during the visit and changed the conversation from the communication of facts to a discussion of values and preferences. The pre-visit model also facilitated implementation of the decision choice at the time of the visit by allowing the patient to arrive at their appointment prepared to make a cancer screening decision. Patients receiving a DA after their appointment had to communicate their screening choice with their provider at a second appointment or through other follow up channels. Regardless of timing or method of distribution, nearly all patients indicated DAs were helpful in their decision making (86– 96%) and would recommend them to other patients (94–100%). 4. Discussion and conclusion 4.1. Discussion When encouraging adoption of a desired behavior, in this case the use of DAs, inertia is a barrier to overcome. It has been shown that default (automatic) options are considerably more likely to be chosen than options requiring some initiative, however minimal [11]. Our results show that automatic delivery of DAs led to a high rate of reaching eligible patients, while distribution strategies that required the initiative of the patient or their physician led to low rates. While method 3 may seem automatic, it relied on the initiative of medical assistants who were distracted by other duties and were only able to screen for eligibility and distribute DAs to a minority of patients. A system that determines eligibility at the time of the visit may be simpler to administer, but does not allow the advantages of the patient viewing the DA before the visit. When the patient arrives at the visit having viewed the DA, the conversation with the clinician is richer, decision making is more efficient and effective, and the decision choice can be implemented at that time. Identifying all patients eligible for cancer screening and delivering DAs to them before a visit presents logistical challenges and requires more resources. In a model system, data from an electronic medical record and computerized scheduling data could be merged to efficiently identify eligible patients before the visit, and avoid the problem of distributing DAs to patients meeting exclusion criteria.
168
C. Brackett et al. / Patient Education and Counseling 78 (2010) 166–168
4.2. Conclusion
Acknowledgements
An automatic DA distribution method is more effective than relying on individual initiative. Enabling patients to view the DA before the visit is preferred.
We are grateful to Martha Coutermarsh, RN, Katrina Whyman, Inger Imset, Ephrem Micaiah, MPH and the primary care service of the WRJVA for their work on this project.
4.3. Practice implications
References
Cancer screening DAs should be systematically offered to all eligible patients before the visit with their clinician. However, this presents logistical challenges in getting DAs to the right patients at the right time. Role of funding The Foundation for Informed Medical Decision Making provided funding for this research, and had no involvement in the study design, collection, analysis and interpretation of data, writing of the report, or decision to submit for publication. FIMDM developed the decision aids used in the study. Conflict of interest The authors have no conflicts of interest, other than receiving funding as above.
[1] Screening for prostate cancer: US Preventive Services Task Force Recommendation statement. Ann Intern Med 2008;149:185–91. [2] American College of Physician. Screening for prostate cancer. Ann Intern Med 1997;126:480–4. [3] American Academy of Family Physicians. Appendix F, summary of policy recommendations for periodic health exam. In: AAFP reference manual. Kansas City, MO: AAFP; 1998. p. 62. [4] American Urological Association. Prostate specific antigen best practice policy. Oncology 2000;14:267–80. [5] US Preventive Services Task Force. Screening for colorectal cancer: recommendation and rationale. Ann Intern Med 2002;137:129–31. [6] Barry MJ, Mulley AG, Fowler FJ, Wennberg JW. Watchful waiting vs. immediate transurethral resection for symptomatic prostatism. JAMA 1988;259:3010–7. [7] Barry M. Health decision aids to facilitate shared decision making in office practice. Ann Intern Med 2002;136:127–35. [8] O’Connor A, Legare F, Stacey D. Risk communication in practice: the contribution of decision aids. BMJ 2003;327:736–40. [9] O’Connor A, Rostom A, Fiset V, Tetroe J, Entwistle V, Llewellyn-Thomas H, et al. Decision aids for patients facing health treatment or screening decisions: systematic review. BMJ 1999;319:731–4. [10] Yarnall K, Pollack K, Ostbe T, Krause K. Primary care: is there enough time for prevention? Am J Public Health 2003;93:635–41. [11] Thaler RH, Sunstein CR. Nudge: improving decisions about health, wealth, and happiness. Caravan Books; 2008.