Facilitators and Barriers to Survey Participation by Physicians

Abstract

Surveys of health care providers are a well-established tool for obtaining information about the organization and delivery of care as well as about provider knowledge and attitudes. However, declining response rates to provider surveys are a widely acknowledged concern. Although a number of studies have identified specific methods for increasing response rates in health care provider—and particularly physician—surveys, few have addressed the more fundamental question of what motivates or deters providers from survey participation. We briefly review theoretical perspectives concerning why providers choose to participate in surveys, and what is known about facilitators and barriers to participation. We then describe several research designs (i.e., focus groups, key informant interviews, diary and office workflow studies, surveying the surveyors, and follow-back studies of respondents/nonrespondents) for obtaining empirical data on facilitators and barriers to survey participation, particularly by physicians and medical groups. Researchers must begin to build an evidence base for understanding provider decisions concerning survey participation.

Keywords

survey physician response rates qualitative research clinicians

Introduction

Surveys of health care providers are a well-established and widely used tool for obtaining information about clinical care practices, the structures and processes for improving care that providers may be using, and the organizational and community settings in which health care is delivered. In an era in which the health care system in the United States is undergoing rapid change—driven in part by health care reform efforts at the national, state, and local levels—gathering information on the experiences and perspectives of providers through surveys will be especially important. Surveys are often the only practical way to obtain information from large and diverse populations of providers on such questions as (1) to what extent are providers using new tools for evaluating and improving the care of their patients (i.e., disease registries, internal quality measurement, decision support systems, nurse care managers for patients with chronic illnesses, etc.)? (2) what are the incentive and reimbursement systems under which providers function? To what extent are clinicians paid on the basis of the volume, quality, and/or cost of the services they provide, and/or on patient experiences? (3) Do the organizational and care management innovations believed to be fostered by new entities such as patient-centered medical homes and accountable care organizations actually affect clinical care delivery? Answers to these and related questions are valuable in their own right and can be even more so when they are linked to other data sources, such as Medicare claims, that may provide objective measures of quality and cost.

However, health care providers—particularly physicians and medical groups—are widely regarded as challenging populations to survey, and these challenges appear to be increasing. Survey researchers experience serious difficulties contacting practicing physicians and identifying appropriate respondents within medical groups (Klabunde et al., 2012); these difficulties may be exacerbated by recent and accelerating trends toward larger practice size and corporate ownership of medical practices. It has been suggested that surveys of physicians generally attain response rates that are 10 or more percentage points lower than those of the general population (Cummings, Savitz, & Konrad, 2001). Further, although response rate is not necessarily a reliable measure of data quality (Johnson & Wislar, 2012), it remains of considerable importance, especially given the recent evidence that response rates for surveys of individual health care providers may be declining (Cook, Dickinson, & Eccles, 2009; Cull, O’Connor, Sharp, & Tang, 2005; McLeod, Klabunde, Willis, & Stark, 2013). For example, response rates have steadily declined during four sequential surveys of the U.S. medical groups conducted between 2000 and 2013 as a component of the National Survey of Physician Organizations (Lawrence P. Casalino, February 20, 2013, personal communication).

A number of studies have been conducted to identify specific methods for increasing participation in health care provider—and particularly physician—surveys. Factors that have been examined include the mode of contact and administration, questionnaire design, incentive type and amount, and follow-up contacts with nonrespondents (Kellerman & Herold, 2001; VanGeest & Johnson, 2011; VanGeest, Johnson, & Welch, 2007). However, few studies have addressed the more fundamental question of what motivates or deters providers from survey participation. In this article, we briefly review what is known about facilitators and barriers to survey participation by health care providers, guided by existing theoretical perspectives on why providers may choose or refrain from survey participation. We then describe several research designs for obtaining much-needed empirical data on facilitators and barriers to survey participation, particularly by physicians and medical groups. We focus on survey participation by these respondent types for two reasons (1) physicians and physician groups play a central role in health care delivery and in current health reform efforts and (2) physicians practice in a wide variety of settings, including private offices, hospitals, community health centers, and nursing homes—different practice settings may present unique challenges to survey design and administration. We leave a more detailed discussion of nonphysician clinicians and nonoffice-based practice settings as a topic for a separate, future paper.

Our article is intended as a “call to action” for researchers to begin building an evidence base that will lead to better understanding of reasons behind provider decisions to participate, or not, in surveys. This new knowledge can then be used to improve the design, conduct, and quality of health care provider surveys.

Theories of Survey Participation

Most research in survey methodology has involved empirical investigations of factors that influence survey participation generally but without an underlying theory as a guide. An attempt to fill this void has been provided by Albaum and Smith (2012) who review theories of why respondents choose to participate in general population surveys. They describe four categories of theories centered on respondent motivation (a) social exchange that focuses on the relative balance between reward and cost; (b) cognitive dissonance, which posits that respondents participate to avoid negative feelings associated with nonresponse; (c) self-perception, which concerns whether the respondent considers himself or herself to be the type of individual who participates in a survey; and (d) commitment/ involvement, where respondents are considered likely to respond to the extent that they view the topic, sponsor, or survey researcher as relevant to themselves (see Gideon, 2012 for more details). An alternative, more encompassing theoretical perspective was given by Cialdini (1988); Groves and Couper (1988), and Groves, Cialdini, and Couper (1992) who specify six psychological concepts, or principles, which survey practitioners should take into account when designing a survey administration system: (1) reciprocation; (2) consistency; (3) social validation; (4) authority; (5) scarcity; and (6) liking. In brief, reciprocation refers to a social norm directing individuals to “return the favor” and has been used to explain the positive effects of survey incentives. Consistency is the tendency to behave in a manner consistent with past behavior, as when agreeing to participate continually in a repeated (panel) survey. The principle of social validation suggests that respondents will be likely to participate when they perceive that others like them have done so. Authority concerns respondents’ perception of the legitimacy or reputation of the surveying organization and perhaps also their view of the utility of the questionnaire instrument. Scarcity applies when potential respondents are inclined to participate when they view that opportunity to be limited and therefore to have more value. Finally, liking describes the degree to which the individual approached to complete a survey is positively inclined toward the person making the request.

Groves et al. (1992) discuss the practical implications of these principles, but mainly in the context of interviewer-administered survey questionnaires, where the practice of tailoring the behavior of the interviewer to the particular respondent can be incorporated. As the authors admit, however, extensions to the case where self-administration is used—and no interviewer exists—may reduce the potential usefulness of a tailoring approach. Hence, the application of this framework to physician surveys, which increasingly make use of mailed and Internet-based self-administration (McLeod et al., 2013), is less clear.

As a further theoretical approach to survey participation generally (i.e., for the general population), Groves, Singer, and Corning (2000) introduced the Leverage-Saliency Theory of Survey Participation, which extends the concept of tailoring by specifying mechanisms by which characteristics of household members influence various survey design features. In brief, the theory proposes a threshold model of the individual’s survey response propensity, where decisions to participate depend on leverage that is applied by the survey researcher to the potential respondents’ critical leverage points. For example, if a particular individual is prone to respond based on incentives provided—his leverage point—then the researcher can make that feature salient by focusing on that aspect (e.g., by providing an incentive and emphasizing that fact). If a different potential respondent can be described as having a leverage point related to feelings of civic duty, then successful participation can be obtained by making that element salient (e.g., through a statement that survey completion is consistent with citizenship).

To apply these concepts, it is necessary to determine the individual’s (or population subgroup’s) leverage point—often only possible through tailoring by experienced and perceptive interviewers. Again, this theory applies mainly to interviewer administration, although the general notion of establishing leverage points may be useful within physician surveys, to the extent that designers can determine which levers must be “pushed,” and what types of messages or interventions are successful in this regard. Figure 1 portrays what Albaum and Smith (2012) term a decision-stage model, outlining the series of steps in the process for completing a typical physician survey; each step is a potential leverage point for encouraging response to the survey. In our following discussion, we make reference to these steps, as each may influence the overall likelihood of participation in a physician survey.

Figure 1.

Decision-stage model of the steps required for successful participation in a physician survey.

Evidence on Facilitators and Barriers to Survey Participation by Health Care Providers

Understanding facilitators and barriers to participation in surveys by health care providers has been a topic of considerable interest to researchers, and there is a sizable body of literature on evaluation of specific methods for facilitating response, particularly in surveys of physicians. This literature has been synthesized in several published reviews. Below, we briefly touch on main findings from these reviews and comment on gaps in the evidence base.

In a classic and much-cited review article describing approaches to improving cooperation in surveys of professionals, including physicians, Sudman (1985) posited that the target population must perceive the survey as having professional relevance, the questionnaire to focus on a single topic, and the survey results to have benefit beyond the researcher’s interests and/or career advancement. Sudman’s findings were reinforced by a subsequent Cochrane systematic review of randomized controlled trial studies to enhance response rates in health-related surveys; the review identified “interesting questionnaire topic” as one of the most effective strategies for achieving higher response rates (Edwards et al., 2009). This finding possibly speaks to the importance of the Groves et al.’s (1992) principle of authority. More recently, several reviews of health care provider survey methods have been conducted, which have had a somewhat narrower focus on methods that survey researchers can use to encourage response; these are summarized in the next paragraph.

With respect to the mode of contacting respondents, mail has been shown to be more effective than e-mail (McLeod et al., 2013), and priority mail to be more effective than standard mail in enhancing response rates in health care provider surveys (Edwards et al., 2009). As noted by Sudman (1985), VanGeest, Johnson, and Welch (2007), and others, it is important to establish the survey’s relevance and prenotification letters and sponsorship or endorsement by organizations salient to the survey’s target population—in a way that promotes authority—have been shown to be facilitators (Edwards et al., 2009; Field et al., 2002; McLeod et al., 2013). In terms of mode of survey administration, studies have shown that surveys using mail, telephone, or mixed modes achieve higher response rates compared with e-mail or Internet-only surveys (McLeod et al., 2013; VanGeest et al., 2007). There is some evidence that shorter questionnaires are associated with higher response rates, particularly among physicians (Edwards et al., 2009; Kellerman & Herold, 2001; VanGeest et al., 2007). Incentives have been a much studied aspect of health care provider surveys, and the weight of the evidence indicates that prepaid monetary incentives are effective in improving response rates, particularly among physicians (Field et al., 2002; Kellerman & Herold, 2001; McLeod et al., 2013; Thorpe et al., 2009; VanGeest et al., 2007). Less studied has been the effectiveness of follow-up contacts with nonrespondents, although use of follow-up was affirmed in one review (Edwards et al., 2009) and the importance of multiple contacts and longer field periods in physician surveys noted in another (VanGeest et al., 2007). Results from a review of nurse surveys suggest that the above-mentioned findings regarding specific approaches to mode of contact and administration, questionnaire design, incentives, and follow-up contacts appear to be generally applicable to nurses, the largest segment of the health care provider workforce (VanGeest & Johnson, 2011).

It is important to note, though, that there is virtually no published literature that addresses variation across provider type or that focuses on the influence of physician specialty. Further, methods have not been developed for successfully implementing surveys of medical groups. Medical groups pose a particular challenge to survey researchers, for several reasons (1) defining the appropriate unit of analysis may not be straightforward, as groups can have multiple sites and levels (i.e., a large medical group can own several smaller clinics); (2) identifying and accessing the appropriate respondent within the group may be difficult; and (3) a single, comprehensive national database of the U.S. medical groups that could be used as a survey sampling frame does not currently exist (Klabunde et al., 2012). Moreover, in both medical group and physician office settings, there is the special challenge of gatekeepers who often control access to the health care providers that survey researchers wish to reach (see Figure 1). Gatekeepers are generally front office staff who greet patients and answer telephones, and they tend to have relatively low status, high turnover, little background in or understanding of research, and may have specific directives to refuse survey requests. Often regarded as integral to the success of a health care provider survey but also as a barrier to obtaining providers’ participation, gatekeepers are relatively unstudied.

In contrast to survey participation facilitators, there is considerably less empirical literature on barriers to survey participation by health care providers. A recent study was motivated by a very low response rate (14%) to a survey of physicians in British Columbia (Wiebe, Kaczorowski, & MacKay, 2012). The researchers faxed brief follow back questionnaires to all nonrespondents. A novel and unexpected finding was that one third of the nonresponding physicians had an office policy to not participate in any surveys. The authors note that the use of evidence-based strategies involving survey administration materials to increase response rates would not work with such physicians because the no-survey policy would preclude them from seeing any of the materials addressed to them. Another study addressed the role of the interviewer in gaining cooperation from gatekeepers to reach physicians for potential survey participation; interviewers’ experience as opposed to their demographic characteristics was found to be important in reducing gatekeeper resistance (Parsons, Johnson, Warnecke, & Kaluzny, 1993). Finally, in his review, Sudman (1985) identified four main reasons for professionals’ refusal to participate in surveys; these included professionals’ (1) lack of time; (2) perception of the survey as having low value; (3) concerns about the confidentiality of results; and (4) perception of questionnaire items as being biased or not providing a full range of responses. We found no empirical literature on barriers to participation of medical groups in surveys.

Research Designs for Obtaining Empirical Data on Facilitators and Barriers to Survey Participation

In a recent article (Klabunde et al., 2012), we used information gleaned from a review of 117 health care provider surveys published during 2000–2010 and a survey methods workshop convened by the National Cancer Institute in 2010 to propose a research agenda for improving the quality of surveys of physicians and medical groups. Among the research questions that we identified where there is great need for more information are:

What motivates health care providers to either participate in surveys or to decline?

What is the nature of the organization and flow of the office practice environment and especially the roles and responsibilities of gatekeepers within the practice?

Why do providers—particularly physicians—react either positively or negatively to specific survey modes? For example, researchers do not have a good grasp of why physicians are less likely to respond to Internet surveys than to paper-based mail surveys.

Should incentives be provided, and if so, how and to whom? This issue is especially pertinent for surveys of medical groups where responses from multiple individuals within the practice may be desired.

What overall burden on providers is imposed by the survey industry at large? Specifically, what volume of survey requests is received by physician offices and the typical practicing physician?

Below, we describe several research designs that might yield much-needed empirical data to address these and other areas of information need in the health care provider survey field. We pay particular attention to qualitative research designs, as it is our belief that qualitative investigations are underused and could provide important insights regarding facilitators and barriers to survey participation, particularly by physicians and medical groups.

Focus Groups

Focus groups are a qualitative data-gathering technique in which an interviewer asks a group of people about their perceptions, beliefs, and experiences related to a particular topic (Fontana & Frey, 1994). This method is potentially valuable in that it provides an efficient means for obtaining a range of opinions simultaneously from a group of individuals and determining which ideas appear to be dominant. However, convening a group of practicing physicians may be challenging, given the physicians’ limited time and scheduling constraints and potential reluctance to participate if they do not view the activity as having value. Further, there is a danger of self-selection bias, as researchers may only be successful in recruiting physicians to focus groups who would also be positively inclined to participate in a survey. There are, though, contexts or environments in which physician focus groups may be feasible, such as part of a Continuing Medical Education course or at a professional association meeting where they to some extent are a captive audience. In one example, the Centers for Disease Control and Prevention has successfully used focus groups of physicians convened by telephone to inform the design of surveys on cancer screening (Henderson et al., 2011). The use of virtual technology (e.g., web conferencing or webinars) also may be effective in reducing barriers to physician participation by lessening the logistical difficulties associated with assembling a group.

Focus groups of gatekeepers such as medical office staff might also be illuminating, especially in discussing and clarifying the pressures associated with their professional activities that may reduce the chances that they specifically convey a survey request to a physician. Further, such groups may provide useful information concerning the ways in which information flows within the practice, and how survey requests might be introduced effectively within that flow. A potential drawback to this design is that gatekeeper staff may have limited control over their own time and might not be able to stop work to participate in a focus group or Internet-based discussion. A financial incentive may be required to encourage participation.

Key Informant Interviews

Key informant interviews can be considered a special type of qualitative or cognitive interview (Willis, 2005), where an individual rather than a group interviewing format is used. Such interviews have successfully been conducted for physicians—such as by Smith, Mingay, Jobe, and Weed (1992)—in a study to determine how physicians cognitively process information contained in death certificates. Key informant interviews of clinical and administrative staff have also been used in the National Survey of Physician Organizations to examine the complexity of organizational cultures present in U.S. medical groups (Nembhard, Singer, Shortell, Rittenhouse, & Casalino, 2012). The conduct of individual interviews of office staff, or other gatekeepers, could be particularly useful but may again be challenging unless these are scheduled for nonwork hours. Key informant interviews can be conducted in person or by telephone, and they are less subject to some of the disadvantages of the focus group format, such as domination of the group by one person, the potential for “group thinking,” and participant reluctance to discuss sensitive or controversial topics in a group setting (Fontana & Frey, 1994). Nevertheless, whether the format of the interview involves a group or individual context, several key issues could be addressed through qualitative interviews, including the following:

Clinician receptiveness to surveys: In general, what factors influence the decision to either establish a general policy concerning surveys or to the extent survey requests are even considered, to participate in a specific survey.

What elements do physicians focus on when reviewing a survey request: the identity of the sponsoring organization; their perceptions of the importance of the topic; the incentive provided or promised; their perception of the burden involved; or something else?

How survey requests are handled within the clinician’s practice setting; whether or not a survey request even reaches the physician; and the role of gatekeepers in directing the survey request, and/or the questionnaire, to the appropriate respondent.

The frequency with which survey requests arrive at the physician office, what form they take (e.g., letters, e-mail, phone call, fax); and from what types of organizations (public vs. private, and especially whether these are from industry, such as pharmaceutical companies).

Whether physicians or other staff believe that there are steps that survey organizations can take to increase the likelihood that clinicians will choose to participate.

Diary Studies

In order to obtain more systematic information concerning burden and load of surveys received, clinicians and gatekeepers could be asked to keep track, over a defined period of time, of the number of survey requests received and detailed information concerning the topic, sponsoring organization, whether they participated in the survey or not, and why. This approach may be most effective when carried out by members of a panel specifically recruited for this type of study. We recognize, however, that threats to the validity of this approach include panel attrition, and self-selection of participants who could introduce significant bias. Collection of information might be facilitated by use of personal digital assistant (PDA) devices in which a diary application provides a structured format for responses and also periodically prompts clinician participants to record diary entries.

Office Workflow Studies

It has been argued that researchers tend to focus on narrow topics with targeted samples, and that they no longer seek to establish “intimate familiarity with research participants and their worlds” (Charmaz, 2004). Office workflow studies might provide survey researchers with a better understanding of how a physician practice or medical office functions and therefore insights into optimal ways of directing survey requests and packets to health care providers. An office workflow assessment can be considered a type of participant observation study that could prove useful for exploring the processes and behaviors occurring in the clinical practice environment (Adler & Adler, 1994). The potential drawbacks to a participant observation approach are that there is tremendous heterogeneity in the size and organization of medical practices in the United States; results from a workflow study conducted in only a few practices might not be very generalizable. Second, given the physical constraints and fast pace of a typical medical office as well as the need to maintain patient privacy and confidentiality, it might be more practical to ask physicians and gatekeepers to describe processes and procedures in their offices as opposed to trying to observe this in “real time.” Nevertheless, we believe that office workflow studies might be beneficial in informing the design and conduct of health care provider surveys.

Survey the Surveyors

As evident from the published literature, surveys of health care providers—both large and small scale—are quite common (Cook et al., 2009; McLeod et al., 2013; VanGeest & Johnson, 2011; VanGeest et al., 2007). A survey of the research organizations that have conducted recent surveys might yield valuable information about facilitators and barriers to implementing health care provider surveys. For example, a lead organization could take responsibility for obtaining information from surveyors, especially pertaining to their experiences with respect to (a) trends and changes over time in survey procedures (e.g., use of incentives, type provided, and amount); (b) best practices for surveying care providers (with respect to mode, combination of mode, or mode switching); and (c) development of novel approaches in general. Data collection could be either standardized, as through an Internet survey, or a semistructured format such as that applied in focus groups, or key informant interviews could be especially efficient and informative. Although a potentially novel approach, a clear potential drawback to surveying the surveyors is that survey research organizations may be reticent to share information that they consider to be proprietary.

Follow Back Studies of Responders and Nonresponders

Follow back to nonrespondents is used in general population surveys both to investigate the characteristics of those who refuse and to assess their reasons for nonparticipation (Hansen & Hurvitz, 1946; Stoop, 2004). Normally follow back requires the expenditure of significant additional resources (time, incentives) beyond the initial survey request, as the intended respondents are by definition resistant to survey participation. We have already mentioned the follow-back study of nonrespondents to a physician survey conducted by Wiebe and colleagues (2012) as one example specific to health care provider surveys. An even less common approach, but one that we believe is vital for understanding participation as well as nonparticipation, is to return to survey respondents who have completed the questionnaire to ask them why they did choose to comply with the survey request. That is, what features of the overall survey design and administration system compelled respondents to participate? It is likely that in any follow-back study of nonrespondents or respondents, the survey would need to be very brief and focused specifically on barriers/facilitators to survey participation; access to and cooperation of gatekeepers would also be required, and consideration should be given to providing an incentive as well as to making multiple follow-up contacts.

Conclusions

Minimizing errors of nonobservation, such as those due to nonresponse, as well as those of observation—particularly response error—is a central tenet of survey research (Groves, Fowler, Couper, Ledowski, & Singer, 2009). Yet, there is a paucity of empirical studies to address the fundamental question of what motivates or deters health care providers from participating in surveys. Few studies of facilitators and barriers to health care provider participation in surveys have been conducted, and we know little about how facilitators and barriers may differ by specialty. We described several research designs—including focus groups, key informant interviews, diary and office workflow studies, surveying the surveyors, and follow-back studies of respondents/nonrespondents—that could be employed to obtain much-needed empirical data on facilitators and barriers to survey participation, particularly by physicians and medical groups. Survey researchers should look for opportunities to undertake such studies, either as stand alone projects or integrated into the design of a larger survey effort. A mixed-methods approach to studying facilitators and barriers to health care provider survey participation, such as a follow-back study coupled with key informant interviews, might provide a particularly rich context for understanding providers’ motivations to participate—or not—in a given survey.

We also noted that very little of the published work evaluating methods for improving health care providers’ survey participation appears to be theory based. It is therefore unclear whether extant theories of survey participation developed for the general population can be applied effectively to health care providers. Some of the elements described by Groves and colleagues (Groves, Cialdini, & Couper, 1992; Groves & Couper 1988; Groves, Singer, & Corning, 2000) may be quite relevant to physicians as respondents. For example, the concept of reciprocation supports the finding that physicians, as with general population respondents, can be motivated to respond by monetary incentives. However, there are other influences on providers that may not be well captured by existing theories. For one, the key role of gatekeepers in the survey administration process requires an approach that recognizes the shared or distributed nature of facilitators and barriers and may need to take into account a diversity of influences that impact at multiple lever points. Further, there may be factors that are somewhat unique to provider surveys, such as the existence of office policies that deter survey participation. To some extent, then, a survey of health care providers may more closely resemble a business or establishment survey than that of individuals and require a less individualistic theoretical perspective.

However, even at the level of the individual clinician’s decision making, there may be factors that differ from those which are more salient to general population respondents. Clinicians may perceive an element of threat to answering survey questions, either because this may compromise business-relevant information or because they fear that the responses may be used in a way that reflects negatively on their practice or profession (e.g., reporting by the survey organization that physicians fail to adhere to supposed best practices or lack key medical knowledge). A comprehensive theory of participation in health care provider surveys would optimally take into account such potential influences. The qualitative study designs that we described, including focus groups and key informant interviews, might have particular utility in informing a health care provider-specific theory of survey participation.

Finally, the absence of studies to examine facilitators and barriers to conducting surveys of medical groups is especially noteworthy. This type of survey is of growing importance because of the changes underway in the organization and delivery of health care in the United States, catalyzed in part by the passage of national health care reform legislation in 2010. We view the empirical study of facilitators and barriers to conducting surveys of medical groups as a wide open area of investigation and encourage survey researchers to consider applying the study designs that we described in this article as well as to think creatively about other ways to contribute to the knowledge base in this area.

In summary, surveys of health care providers—both individual clinicians and groups of care providers—are an essential tool for obtaining information about clinical care practices, influences on care delivery, and the diverse settings in which health services are rendered. However, response rates to such surveys appear to be in decline, and survey researchers report increasing difficulties in securing providers’ participation. A concerted effort by survey researchers to build both an evidence base and a theoretical perspective for understanding provider decisions concerning survey participation would greatly help to inform future efforts in designing and administering health care provider surveys of the highest quality and utility.

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.

References

Adler

P. A.

Adler

(1994). Observational techniques. In Denzin

N. K.

Lincoln

Y. S.

(Eds.), Handbook of qualitative research (pp. 377–392). Thousand Oaks, CA: Sage.

Albaum

Smith

S. M.

(2012). Why people agree to participate in surveys. In Gideon

(Ed.), Handbook of survey methodologys for the social sciences (pp. 179–193). New York, NY: Springer.

Charmaz

(2004). Premises, principles, and practices in qualitative research: revisiting the foundations. Qualitative Health Research, 14, 976–993. doi:10.1177/1049732304266795

Cialdini

R. B.

(1988). Influence: Science and practice. Glenview, IL: Scott, Foresman.

Cook

J. V.

Dickinson

H. O.

Eccles

M. P.

(2009). Response rates in postal surveys of healthcare professionals between 1996 and 2005: An observational study. BMC Health Services Research, 9, 160. doi:10.1186/1472-6963-9-160

Cull

W. L.

O’Connor

K. G.

Sharp

Tang

S. F.

(2005). Response rates and response bias for 50 surveys of pediatricians. Health Services Research, 40, 213–226. doi:10.1111/j.1475-6773.2005.00350.x

Cummings

S. M.

Savitz

L. A.

Konrad

T. R.

(2001). Reported response rates to mailed physician questionnaires. HSR: Heath Services Research, 35, 1347–1355.

Edwards

Roberts

Clarke

M. J.

Diguiseppi

Wentz

Kwan

Cooper

Felix

L. M.

Pratap

. (2009). Methods to increase response to postal and electronic questionnaires. Cochrane Database Systematic Review, 3, MR000008. doi:10.1002/14651858 MR000008.pub4

Fontana

Frey

J. H.

(1994). Interviewing: The art of science. In Denzin

N. K.

Lincoln

Y. S.

(Eds.), Handbook of qualitative research (pp. 361–376). Thousand Oaks, CA: Sage.

10.

Field

T. S.

Cadoret

C. A.

Brown

M. L.

Ford

Greene

S. M.

Hill

D.,

Zapka

J. M.

(2002). Surveying physicians: Do components of the “Total Design Approach” to optimizing survey response rates apply to physicians? Medical Care, 40, 596–605.

11.

Gideon

(2012). Handbook of survey methodologys for the social sciences. New York, NY: Springer.

12.

Groves

R. M.

Cialdini

R. B.

Couper

M. P.

(1992). Understanding the decision to participate in a survey. Public Opinion Quarterly, 56, 475–495.

13.

Groves

R. M.

Couper

M. P.

(1988). Nonresponse in household interview surveys. New York, NY: Wiley.

14.

Groves

R. M.

Fowler

F. J.

Couper

M. K.

Ledowski

J. M.

Singer

(2009). Survey methodology (2nd ed.). Hoboken, NJ: John Wiley.

15.

Groves

R. M.

Singer

Corning

(2000). Leverage-salience theory of survey participation: Description and an illustration. Public Opinion Quarterly, 64, 299–308.

16.

Hansen

M. H.

Hurwitz

W. N.

(1946). The problem of nonresponse in sample surveys. Journal of the American Statistical Association, 41, 517–529.

17.

Johnson

Wislar

(2012). Response rates and nonresponse error in surveys. The Journal of the American Medical Association, 307, 1805–1806.

18.

Henderson

DeGroff

Richards

T. B.

Kish-Doto

Soloe

Heminger

Rohan

(2011). A qualitative analysis of lung cancer screening practices by primary care physicians. Journal of Community Health, 36, 949–956. doi:10.1007/s10900-011-9394-2

19.

Kellerman

S. E.

Herold

(2001). Physician response to surveys: A review of the literature. American Journal of Preventive Medicine, 20, 61–67. doi:10.1016/S0749-3797(00)00258-0

20.

Klabunde

C. N.

Willis

G. B.

McLeod

Dillman

D. A.

Johnson

Greene

S. M.

Brown

M. L.

(2012). Improving the quality of surveys of physicians and medical groups: A research agenda. Evaluation and the Health Professions, 35, 477–506.

21.

McLeod

C. C.

Klabunde

C. N.

Willis

G. B.

Stark

(2013). Health care provider surveys in the U.S., 2000-2010: A review. Evaluation and the Health Professions, 36, 107–127. doi:10.1177/0163278712474001

22.

Nembhard

I. M.

Singer

S. J.

Shortell

S. M.

Rittenhouse

Casalino

L. P.

(2012). The cultural complexity of medical groups. Health Care Management Review, 37, 200–213. doi:10.1097/HMR.0b013e31822f54 cd

23.

Parsons

J. A.

Johnson

T. P.

Warnecke

R. B.

Kaluzny

(1993). The effect of interviewer characteristics on gatekeeper resistance in surveys of elite populations. Evaluation Review, 17, 131–143. doi:10.1177/0193841X9301700201

24.

Smith

A. F.

Mingay

D. J.

Jobe

J. B.

Weed

J. A.

(1992). A cognitive approach to mortality statistics. In Proceedings of the Section on Survey Research Methods, American Statistical Association, 812–817.

25.

Stoop

(2004). Surveying non-respondents. Field Methods, 16, 23–54.

26.

Sudman

(1985). Mail surveys of reluctant professionals. Evaluation Review, 9, 349–360.

27.

Thorpe

Ryan

McLean

S. L.

Burt

Stewart

Brown

J. B.

Reid

G. J.

Harris

(2009). How to obtain excellent response rates when surveying physicians. Family Practice, 26, 65–68.

28.

VanGeest

J. B.

Johnson

T. P.

Welch

V. L.

(2007). Methodologies for improving response rates in surveys of physicians: A systematic review. Evaluation and the Health Professions, 30, 303–321. doi:10.1177/0163278707307899

29.

VanGeest

J. B.

Johnson

T. P.

(2011). Surveying nurses: Identifying strategies to improve participation. Evaluation and the Health Professions, 34, 487–511. doi:10.1177/0163278711399572

30.

Wiebe

E. R.

Kaczorowski

MacKay

(2012). Why are response rates in clinician surveys declining? Canadian Family Physician, 58, e225–e228.

31.

Willis

G. B.

(2005). Cognitive interviewing: A tool for improving questionnaire design. Thousand Oaks, CA: Sage.