Background

Prostate cancer is the most common non-cutaneous cancer among men in the United States and disproportionately affects Black men; [1, 2] who have the highest prostate cancer incidence and mortality rates [3]. Prostate-specific antigen (PSA) screening is a preference-sensitive decision where harms and benefits can vary significantly across patients. The United States Preventive Services Task Force (USPSTF) Prostate Cancer Screening Statement, as well as the American Urological Association (AUA), the Society of Urologic Oncology (SUO), and the American Cancer Society (ACS) recommend a shared decision making (SDM) approach, where clinicians and patients discuss the evidence and patients are supported to achieve informed decisions [4,5,6].

In spite of changing PSA-screening recommendations, the use of SDM has not yet been widely adopted [7,8,9]. Men considering prostate cancer screening are rarely encouraged to clarify their preferences or engage in a balanced discussion to weigh the pros and cons of screening [10,11,12]. USPSTF has recognized the underutilization of SDM in practice, especially among Black men, and has emphasized the need to understand how best to implement and adapt SDM programs within diverse populations. In 2018, the USPSTF changed their recommendation for PSA screening from a grade D (recommending against screening) to C (supporting individualized decision making based on a patient’s balance of benefits and harms) for men 55–69 due to new evidence of long-term benefit and of less harm from overdiagnosis. However, in contrast to the ACS and AUA guidelines [5, 6], which recommend earlier screening of high-risk men, including Black men, the USPSTF makes no recommendations regarding earlier screening of high-risk men [13]. Our team performed an assessment of PSA screening discussions to understand whether SDM was occurring under standard care conditions. As previous research on SDM in PSA screening has relied on self-reported [7, 10,11,12], we sought to understand these conversations more directly by audio-recording relevant clinic visits. With the overall aim of conducting a qualitative pilot study to guide future intervention development, the objective of this study is to examine the extent of SDM practice in routine primary care appointments.

Methods

Guided by the indicators of high-quality SDM in the “observing patient involvement in decision making” (OPTION) scoring method [14] and the principles of grounded theory [15], audio-recorded patient-provider interactions were analyzed to appraise the use of SDM in PSA screening decisions. This study was designed to establish a baseline for a larger randomized trial evaluating the efficacy of a Community Health Worker (CHW)-led decision-coaching program [16]. The protocol was approved by the [institution] IRB and written documentation of informed consent was received prior to starting data collection.

Setting and Participants

The study was performed at an urban Federally Qualified Health Center (FQHC) primary care clinic. Patients were eligible for the study if they were: (1) biologically male, (2) Black, (3) between the ages of 40 and 69 years, and (4) attending the FQHC for routine primary care appointments. Patients were excluded if they were: (1) visiting their provider for any indication other than a well-visit; (2) seen within 9 months of other PSA tests; (3) seen within 180 days after primary diagnosis of urinary obstruction, prostatitis, hematuria, other disorder of prostate, unexplained weight loss, or lumbar back pain; or (4) had a prior diagnosis of prostate cancer (ICD-10-CM C61). The adult male patient population was cared for at the time by 5 full-time primary care providers, all of whom participated in this study. Patient participants were a purposive sample of those who scheduled routine visits during the study period (June through October, 2019).

Data Collection

Audio-recorded patient-provider encounters were transcribed, and imported into Nvivo software for qualitative data analyses. All patient visits lasted less than 15 min. Study staff set up recording equipment, but were not present in the consultation room.

As discussion around PSA screening was presumed to rarely occur among this patient population, a prompt was used to increase the likelihood that PSA would be discussed. Patients were prompted about PSA in the context of their primary care visit. Providers were handed a document with a prompt to identify the patient as part of the study and to suggest that prostate cancer screening should be discussed in the session. The prompts did not direct the provider or patient as to the content of the PSA discussion.

We determined, a priori, that patients would be sequentially enrolled with ongoing data analysis until thematic saturation was reached [17]. Thematic saturation was quickly achieved (n = 13) due to limited PSA conversation leading to no additional learning or new meaningful data collected after this point.

Data Analysis

Encounter data were coded using procedures designed to ensure thoroughness and reliability. We used Nvivo software to manage the data and coded it according to the principles of ‘Framework for applied policy research’ [18], which consists of a five-stage process including familiarization, identifying themes, indexing, charting, and interpretation. The general development of codes and themes arose from the data, using the principles of grounded theory [15]. To describe the quality of SDM, the OPTION scoring method was applied to each encounter [14]. To enhance objectivity, two researchers took part in the coding and analysis process. All interviews were independently coded by two researchers (AC, KD), who met with another co-author (PU) to resolve discrepancies. When coding was complete, the quotations with each code were examined, summarized, and grouped together into themes.

The OPTION scale is designed to assess the presence/absence and quality of SDM. The OPTION instrument rates a clinician on 12 SDM behaviors. Each item is rated on a skill level of 0 (behavior not observed) to 4 (behavior is exhibited to a very high standard) for a total score range from 0 to 48, with higher scores indicating higher competencies of SDM. The 12 elements of OPTION include: The clinician (1) draws attention to need for decision making process; (2) states there is more than one way to deal with the identified problem; (3) assesses patient’s preferred approach to receiving information; (4) lists options; (5) explains the pros and cons; (6) explores patient expectations; (7) explores the patient’s concerns; (8) checks patient understood; (9) offers explicit opportunities to ask questions; (10) elicits preferred level of involvement in decision making; (11) indicates need for a decision making (or deferring) stage; and (12) indicates need to review the decision. The mean sum OPTION score was transformed to a scale of 0 to 100 using the formula score/48 × 100 [14]. Due to sample size, the scores were intended as a descriptor of SDM and not to draw statistical conclusions.

Qualitative codes derived from OPTION explored the discussion of (1) provider recommendations for or against a screening test; (2) patient-specific factors contributing to the need for screening; (3) patient knowledge of prostate cancer; (4) the pros and/or cons of screening; (5) patient preferences; (6) next steps if screening is positive; and (7) final decision outcome.

Results

Participants

A total of 16 patients were enrolled with 13 recorded patient-provider encounters, representing 5 unique providers and 13 unique patients. Three patients enrolled in the study were not included in the analysis because two did not complete a clinical visit after enrollment and there was an audio recorder malfunction during one patient-provider encounter. The patients had an average age of 54 (range 40–65) including seven age ≥ 55 years and six age 40–54, with 46% identifying as being married or in a domestic relationship, 54% having only a high school education, and 46% currently employed with pay. The FQHC providers consisted of 2 physicians (providers C, E) and 3 nurse practitioners (providers A, B, D). The providers all identified as female and had an average age of 46 with 4 providers identifying as Black (providers A, C, D, E).

Shared Decision-Making Quality

A summary of patient-provider PSA screening discussions is presented in Table 1. By the end of the encounter, 11 patients (85%) planned to receive some form of screening (including 1 digital rectal exam) while 2 patients (15%) declined screening. Providers recommended screening to 92% (12) of the patients. The primary reason for the proposed recommendation was based on age (62%) and/or race (69%). Only 3 patients were directly asked about their screening preferences. Few patients were asked about their prostate cancer knowledge (2), potential urological symptoms (3), or family history (6).

Table 1 PSA screening interactions between black male patients and their providers during primary care visits at a federally qualified healthcare center
Full size table

Most providers consistently discussed the disadvantages (80% of providers) and advantages (80% of providers) of PSA screening. The most common disadvantages cited by providers were possible false positives (100% of cons discussions and 54% of encounters) and the potential for invasive follow-up procedures (71% of cons discussions and 38% of encounters). The most frequent advantages discussed by providers were ease of screening (67% of pros discussions and 46% of encounters), and the ability to detect cancer earlier (67% of pros discussions and 46% of encounters). On average, the younger 40–54 years old age group had greater SDM discussions with their provider. Clinicians indicated African American men are at higher-risk and that earlier screening is suggested with greater frequency in the 40–54 year-old age group (5/6 participants) compared to the older age group (4/7 participants).

The average patient-level OPTION score was 21.54 (range 0–67); the providers’ average score was 27 (range 1–60) (Table 2). The most frequent, and thorough, noted OPTION elements included when the clinician: (1) draws attention to need for decision making process; (2) states there is more than one way to deal with the identified problem; (4) lists options; (5) explains the pros and cons; and (11) indicates need for a decision making (or deferring) stage.

Table 2 Option score elements present per encounter
Full size table

Themes

In addition to the quality codes established a priori, two major themes emerged to describe the PSA-screening-focused interactions. Detailed below, these two themes are: providers explicitly highlighting the importance of SDM and providers stymieing in-depth discussion. Additional quotations supporting these themes are summarized in Table 3.

Table 3 Thematic quotes from patient-provider encounters
Full size table

Highlighting SDM Importance

A theme emerged among providers (40% of encounters) to not only perform SDM but to explicitly mention the importance of having SDM in PSA screening decisions. Providers emphasized the preference sensitive nature of PSA screening telling patients, “it’s becoming more of a gray area. It used to be at this age order this test, but now it’s …patient preference.” Providers described the premise of SDM to explain, “[patients and providers] have a discussion and in the end, [patients and providers] make the decision together.” Further clarifying that a provider “…can’t just say to you the moment you walk in here, do you want PSA or not? Because you’ll be like, ‘I don’t know, you tell me,’” and explaining “…the whole point [of SDM] is for you to be knowledgeable enough about [PSA screening] so you can make a decision about what to do.”

Stymieing In-Depth Discussion

A second theme emerged showing that when opportunities for SDM arose during patient-provider interactions, providers stymied the potential of in-depth discussion when responding to patient inquiries by providing brief responses and declarative statements that closed the interaction to potential SDM discussion. Many providers limited discussion by using declarative statements with only a confirming follow-up question of “…okay?” at the end (e.g. “I am going to include a PSA test, okay?”). These close ended questions from providers were not followed by inquiries or meaningful feedback from patients. Even direct inquiries by patients regarding PSA screening were sometimes met with a brief uninformative response. One provider, when told by a patient that they, “…thought about [prostate screening] but would like to know what the test is all about.” The provider told the patient, “the test is a simple blood test…. Since I am already doing your annual blood, I’ll just include it.”

Discussion

In our study, the first of its kind to use audio recorded patient-provider conversations to assess the presence and quality of SDM in discussions regarding PSA screening, we found that providers were inconsistent in their counseling approach. While some discussions addressed the advantages, disadvantages, and/or scientific uncertainty of screening, few patients received all elements for SDM. Furthermore, while nearly all providers recommended PSA screening, and most patients received it, very few patients were asked directly about their preferences or their knowledge about PSA screening. OPTION scores were low on average with only 1 of the 5 providers receiving an average score of ≥ 50 (a score of 50 indicates SDM “behaviors are observed and a minimum skill level achieved”). These demonstrated limitations in SDM are consistent with existing literature showing that few men are given the opportunity for SDM during PSA screening [11], and men often do not receive all the information they need to participate in SDM [19, 20]. Our study is an important contribution to and advance over previous work because it employs direct audio recording of the patient-provider interaction, which has similarly been used to evaluate other cancer screening SDM [21], and thus affords some qualitative understanding of these shortcomings.

Our study revealed that, even under ideal circumstances where providers and patients are prompted to discuss PSA screening, there are potential factors influencing the use of SDM beyond the oft-cited barriers to SDM (e.g. limited provider time and belief that SDM won’t influence behavior, patient limited medical literacy) [22]. Specifically, there may be a misunderstanding regarding what constitutes SDM, even among providers who explicitly tell their patients of the need for SDM. A majority of providers stating advantages, disadvantages, and recommendations primarily did so using declarative statements with a confirming question “…okay?” tacked on to the end. This format did not create an opportunity for patients to engage in the conversation or express personal preferences; indeed, no patient in our data set went on to ask any clarifying questions after such a statement. This effect on conversation may also be the consequence of the time constraints of the encounter.

Through the prompting of both providers and patients to discuss PSA screening, participants were likely biased towards potential PSA SDM, as compared to standard care. In spite of this, very limited SDM was noted. Consistent with the literature [10,11,12], this indicates that, even under the best possible circumstances, where both patient and provider are primed to discuss PSA screening, high quality SDM rarely occurs. Thus, suggesting further efforts are needed beyond clinical reminders to effectively encourage the use of SDM for PSA screening.

Given the persistently low rates of SDM demonstrated in this study and in national trends [20], investigating alternate strategies to promote SDM is crucial. In the context of the indisputably limited time available for counseling [22, 23], an intervention such as decision aids or a CHW/other trained staff -led decision coaching program may be a more feasible alternative [24, 25]. In preliminary work, using an American Cancer Society (ACS) decision aid [6] in a community-based setting, we found a CHW to be potentially as effective as a physician for increasing prostate cancer knowledge and decreasing decisional conflict [6, 26, 27]. A CHW-led approach may address barriers to provider-led SDM by providing more time for patients to receive counseling necessary for decision making, as well as overcome patient-level barriers such as language differences and limited medical literacy [22, 23, 28]. In the context of the larger trial of a CHW-led decision-coaching program [16], these results helped establish a baseline of current levels of PSA SDM and provided additional insights on clinic work flow that informed the decision to implement the ACS decision-aid guided CHW PSA SDM coaching session prior to the appointment [16]. This study had several limitations. Foremost is the limited sample size prescribed by our study design prevents the ability to test deductive hypotheses for both patients and providers and perform subgroup analyses. Furthermore, information was not available on the background of the patient-provider relationships or previous PSA discussions. Therefore, elements of SDM that might have occurred at previous encounters could not be captured. Similarly, by focusing on only visits with Black men, we cannot know if clinician behavior would differ based on patient race. Our study has several strengths including the use of recorded patient-provider encounters to review the exact words and nuanced communication rather than relying on self-report, thus reducing the potential for recall and social desirability bias, and allowing for an understanding of the breadth and depth of such discussions. Third-party observation has a high correlation with patient report, however, observers may perceive fewer SDM behaviors than patients; potentially biasing the audio-recording review towards less SDM [29].

Conclusion

SDM is critical to ensuring the quality of PSA screening, a highly preference sensitive process. Our study found limited SDM during PSA screening consultations among Black men and their providers. The counseling that did take place used components of SDM but only inconsistently and incompletely. Efforts are needed to develop strategies to improve SDM for prostate cancer screening in diverse patient populations which will ensure the quality of PSA screening and promote health equity in a population especially vulnerable to morbidity from prostate cancer.

New Contribution to the Literature

This research contributes to the current literature by providing evidence, via audiotaped discussions, confirming previous findings that, during PSA counseling, providers may use components of SDM but inconsistently and incompletely. This study highlights the need to further improve SDM for PSA screening.