Freed GL, Uren RL, Hudson EJ, Lakhani I, Wheeler JRC, Stockman JA, Research Advisory Committee of the American Board of Pediatrics FT. Policies and Practices Related to the Role of Board Certification and Recertification of Pediatricians in Hospital Privileging. JAMA. 2006;295(8):905-912. doi:10.1001/jama.295.8.905
Author Affiliations: Child Health Evaluation and Research Unit, and Division of General Pediatrics, University of Michigan (Drs Freed and Wheeler, and Mss Uren, Hudson, and Lakhani), and Department of Health Management and Policy, University of Michigan, Ann Arbor (Drs Freed and Wheeler); and American Board of Pediatrics, Chapel Hill, NC (Dr Stockman).
Context Privileging involves the granting of permission to perform specific professional activities under the jurisdiction of a governing body's (hospital) authority. In 1951, the Joint Commission on the Accreditation of Hospitals (later renamed the Joint Commission on Accreditation of Healthcare Organizations) was formed to codify the process of hospital assessment. In the early part of the 20th century, a parallel process was being undertaken by the medical specialties to evaluate and recognize competence among physicians through the creation of specialty boards.
Objectives To describe the use of board certification in hospital privileging policies for general pediatricians and pediatric subspecialists and to identify any variation among types of hospitals.
Design, Setting, and Participants A telephone survey between January 1 and June 30, 2005, of privileging personnel among a random, weighted sample of 200 nonspecialty hospitals stratified by teaching status, children's vs general hospitals, freestanding children's hospital vs part of hospital system, and urban vs rural location.
Main Outcome Measures Proportion of hospitals that require board certification at initial privileging or at some point to maintain privileges and recertification to maintain privileges.
Results Of 200 hospitals, 7 hospitals were ineligible because they did not have at least 1 pediatrician on staff. One hundred fifty-nine hospitals completed the telephone interview, resulting in an overall response rate of 82%. A total of 124 (78%) of 159 hospitals did not require general pediatricians to be board certified at initial privileging; however, 111 (70%) did require pediatricians to become board certified at some point during their tenure. Of these 124 hospitals, 52 (42%) did not report a time frame in which certification must be achieved. Forty-nine (43%) of 113 hospitals required pediatric subspecialists to achieve subspecialty certification within a specific time frame.
Conclusions These results raise issues regarding the manner in which board certification is used or not used by hospitals in their efforts to ensure the practice of high-quality care within their institutions. The premise for recertification is the need to assure the public of continued competence of physicians over the course of their professional careers. Increased attention by the public and regulatory agencies regarding patient safety and quality of care will likely have an impact on hospital privileging processes.
Privileging involves the granting of permission to perform specific professional activities under the jurisdiction of a governing body's (hospital) authority. The notion for this process began in 1919 with the formal promulgation of a minimum standard by the American College of Surgeons.1 This first effort to codify a listing of what hospitals should do to ensure patient care and safety included the issue of requirements for granting privileges to physicians.2 The standard stated that members of the medical staff needed to be “competent in their respective fields.”1 However, there was no specificity in how that competence was to be evaluated or measured.
In 1951, the Joint Commission on the Accreditation of Hospitals (later renamed the Joint Commission on Accreditation of Healthcare Organizations [JCAHO]) was formed to codify the process of hospital assessment. The first standards manual was published in 1953 and included the requirement of a physician-credentialing committee within hospitals.3 The manual specifically stated that “personnel of each service shall be qualified by training and demonstrated competence, and shall be granted privileges commensurate with their individual abilities.”3 The manner of implementing this directive was left to the individual hospitals to interpret and operationalize.
In the early part of the 20th century, a parallel process was undertaken by the medical specialties to evaluate and recognize competence among physicians. Specialty boards were created to certify the competence of physicians in specific fields of clinical expertise. The American Board of Pediatrics (ABP) was established in 1933 and created a statement of purpose, “The intent of certification is to provide assurance to the public and the medical profession that a certified pediatrician has successfully completed an accredited educational program and various evaluations, including an examination, and that the individual possesses the knowledge, skills, and experience requisite to the provision of high quality care in pediatrics.”4
For several decades, the certificates granted were valid for the entire professional life of a pediatrician (permanent certificates). However, in 1987, the ABP began issuing time-limited certificates, requiring renewal every 7 years to maintain certification.
Little research exists regarding the overlap of hospital and board efforts to determine the competence of physicians and to assure the public that the medical services they receive are provided by a physician who has met specific criteria. It is unknown how or if hospitals use board certification as a proxy measure of professional competence in their privileging decisions. Furthermore, since the advent of recertification requirements, there has been no research that identifies hospital privileging policies regarding the recognition and/or use of this higher standard. Because there is no uniform governmental mandate or requirement from the JCAHO for hospitals to require physician certification or recertification, significant variation may exist. Such variation may also place hospitals at odds with patient preferences. A recent study5 suggested that patients prefer that board-certified physicians provide care to them and their families.
To better understand the interface of board certification and the policies regarding privileging and to identify any variation among different types of hospitals, we conducted a national survey study of hospital policies and practices related to the recognition and use of board certification and recertification of pediatricians.
We developed a telephone survey designed to be completed in 10 minutes or less. The survey focused on the following descriptive research questions. Do hospitals require board certification for all general pediatricians and pediatric subspecialists at the time of initial privileging? Or, do requirements differ for those who have just completed training vs those who have been in practice for some time? Are hospitals tracking board certification for their pediatricians and how do they verify certification? Do hospitals require recertification for general pediatricians and pediatric subspecialists?
Other variables included in the survey focused on whether the hospitals require pediatricians with permanent or time-limited certificates to recertify, the time frames in which certification must occur, and whether hospitals revoke or deny privileges for failure to recertify.
The interview instrument was pilot tested for clarity and ease of use with representatives from a convenience sample of hospitals within the state of Michigan and revised to clarify potentially ambiguous questions. Pilot surveys were not included in the analyses. The study was approved by the institutional review board at the University of Michigan Medical School, Ann Arbor.
All hospitals that did not provide care to children or that only cared for children with specific conditions (eg, burns) were excluded from the American Hospital Association's Annual Survey of Hospitals roster. Using the remaining 2288 hospitals as our sampling population, the research team selected a random sample of 200 nonspecialty hospitals weighted to provide nationally representative estimates. The sample was stratified by the Council of Teaching Hospitals and Health Systems (COTH) designation (teaching vs nonteaching), membership in the National Association of Children's Hospitals and Related Institutions (NACHRI) (children's hospital vs non–children's hospital), freestanding (freestanding children's hospital vs part of hospital system), and metropolitan statistical area size (urban vs rural [not a metropolitan statistical area]).
Hospitals were sampled with varying probabilities from each stratum. Weights were applied to create a representative sample of the overall hospital population. The total sampling weight (TSW) calculated for each hospital was based on the probability of selection into the study (P) and the response rate (RR). The following formula was used: TSW = [1/P] × [1/RR].
Between January 1 and June 30, 2005, research staff attempted to contact the selected hospitals. Interviewers requested to speak with the department responsible for credentialing or privileging at the hospital, typically the office of clinical affairs, the medical staff office, or the physician credentialing/privileging department. When the appropriate person was located, interviewers explained the purpose of the study and obtained verbal consent to participate. A small number of hospital staff requested to complete the information via fax.
During the interview, the respondents were asked scripted questions and either answered yes or no, or supplied a free form response, which was then coded into themes following the complete survey process.
Initially, frequency responses for all survey items were calculated to create descriptive statistics. Then, we performed a cross-tabulation of responses by the various hospital classifications listed above and computed the weighted χ2 statistics for each survey item. SAS version 9.1 (SAS Institute Inc, Cary, NC) was used for all statistical analyses. P<.05 was considered statistically significant.
For purposes of analysis, hospitals that had initial certification requirements that varied for different physicians (eg, hospitals allow recent graduates up to 5 years to obtain their certification but required certification for those physicians in practice for at least 5 years) were classified as having a mixed policy (ie, their certification policy varies for different physicians), requiring certification for some but not all physicians when allowing initial privileges.
Of 200 hospitals, 7 hospitals were ineligible because they did not have at least 1 pediatrician on staff. Of the remaining 193 hospitals, 159 privileging personnel completed the telephone interview, representing an overall response rate of 82%.
Response rates did not differ significantly by COTH status, membership in NACHRI, metropolitan statistical area, or freestanding classification; therefore, there was no impact on the analytic power of the weighting. Because not every hospital responded to every question, the total number for each question may differ slightly due to missing responses.
Of the 159 hospitals that completed the interviews, 68 (43%) were teaching hospitals (COTH), 83 (52%) were members of NACHRI, 35 (22%) were freestanding children's hospitals, and 130 (82%) were located in urban settings. The percentages do not total to 100 because there is overlap among the groupings (ie, a teaching hospital can also be a children's hospital).
Overall, our results showed that 124 hospitals (78%) did not require pediatricians to be board certified at the time of initial privileging; however, 111 hospitals (70%) did require pediatricians to be board certified at some point during their tenure (unweighted data) (Table 1). A variety of hospital policies were implemented, including required certification for all pediatricians at initial privileging; not required certification for any pediatrician at initial privileging but required after some time period; required certification initially for some but not for others (eg, new graduates may be allowed some time to obtain their certification while experienced physicians may be required to be certified initially or the level/category of privileging requested may require certification); or not required certification at initial privileging but required recertification for those physicians who are certified. The last seemingly contradictory policy was a result of hospitals requiring recertification at the intervals JCAHO required assessments of recertification status. Physicians who never had their recertification status assessed as part of a JCAHO assessment were categorized as never certified.
Percentages are unweighted for the total sample; however, all percentages for the COTH vs non-COTH and NACHRI vs non-NACHRI groupings are weighted.
General Pediatricians. One hundred twenty-four hospitals (78%) did not require general pediatricians to be board certified at the time of initial privileging (Table 2). Overall, 29 (18%) of the 159 respondent hospitals stated that their certification policy varied for some types of pediatricians. These hospitals were classified as having a mixed policy for initial certification, which allowed new graduates time to obtain their certification after initial privileging but required experienced physicians to have already obtained certification. However, nearly all of the hospitals that did not require certification initially did require their general pediatricians to have completed residency training. Six respondent hospitals (4%) did require board certification at initial privileging for all general pediatricians. Of these 6 hospitals, 5 (83%) reported that they require the certification to be current.
Of the 124 hospitals that did not require board certification at the time of initial privileging, 114 (92%) required their general pediatricians to at least have successfully completed residency training. The NACHRI hospitals were more likely to require completion of residency than non-NACHRI hospitals did (99% vs 85%, weighted P = .01).
Of the 124 hospitals that did not require board certification, 60 (48%) did report having some time frame in which certification must be achieved. Most of the hospitals that required certification to occur after initial privileging allowed the physicians 4 to 6 years to obtain their certificate. There were no significant differences among the comparison groupings of hospitals.
Overall, 55 respondent hospitals (37%) had exceptions to their general pediatrician certification policies at the time of initial privileging. The NACHRI hospitals were more likely to have exceptions to their policies than non-NACHRI hospitals (44% vs 30%, weighted P = .01).
The most commonly reported policy exceptions for general pediatricians primarily were in 1 of the following 3 domains: grandfather clause (physicians who were hired before the certification policy was in place would not be required to follow the policy); waivers (physicians would not be required to follow the policy if they could demonstrate competency, had a great deal of experience, or were a foreign or international graduate); and case by case (hospitals reported they would work with physicians on a case-by-case or individual basis).
Pediatric Subspecialists. The overall study results showed that 113 (74%) of the 153 hospitals did not require pediatric subspecialists to be board certified at the time of initial privileging (Table 2). However, 103 (91%) of 113 hospitals that initially did not require certification required pediatric subspecialists to have completed fellowship training and 49 hospitals (43%) required pediatric subspecialists to become certified within a specified time frame. Most of the hospitals that required certification to occur after initial privileging allowed the subspecialists 4 to 6 years to obtain their certificate.
Similar to the findings for general pediatricians, 51 respondent hospitals (37%) reported that they do have exceptions to their certification policies for pediatric subspecialists. The NACHRI and non-COTH hospitals were more likely to have exceptions to their policies (Table 3). The most commonly reported policy exceptions for pediatric subspecialists were the same 3 domains as those for general pediatricians: grandfather clause, waivers, and case by case.
Both COTH (11% vs 1%, weighted P<.001) and NACHRI (15% vs 0%, weighted P = .01) hospitals were more likely to have exceptions that varied on a case-by-case basis. However, the non-COTH (60% vs 20%, weighted P = .02) and non-NACHRI (61% vs 20%, weighted P = .02) hospitals were more likely to have exceptions related to a grandfather clause.
In comparison with certified pediatric subspecialists, non–board-certified subspecialists who were credentialed were not often restricted with respect to their privileging. Overall, only 10 respondent hospitals (7%) indicated that they restrict the privileges of non–board-certified pediatric subspecialists. There were no significant differences among the hospital categories.
One hundred thirty-eight respondent hospitals (87%) indicated that they track board certification expiration dates for their pediatricians. Similar results were observed among all hospital categories.
Table 4 describes the methods used by hospitals to verify board certification status. Overall, the most frequently used method was online verification via CertiFACTS (70 [44%]), a certification verification organization that contains data from the 24 American Board of Medical Specialties (ABMS) member boards (http://www.certifacts.org). Significantly more COTH and NACHRI hospitals used CertiFACTS than other hospitals. Some hospitals reported they prefer CertiFACTS because it can be used for all specialties.
One hundred twenty-eight hospitals (82%) reported that they verify pediatricians' certification every 2 years as part of conducting their recredentialing process and as required by the JCAHO. The next most frequent response was to verify certification upon expiration of the certificate (24 hospitals [15%]). There were no significant differences among the comparison hospitals in the frequency of certification verification.
General Pediatricians. Overall, 71 respondent hospitals (45%) stated that they required general pediatricians with time-limited certificates to recertify. There were no significant differences between the hospital groupings. However, of these 71 hospitals, 48 (69%) allowed pediatricians to retain their privileges even after their certificate expired. Again, there were no significant differences between the hospital groupings.
Of the 48 hospitals that allowed pediatricians to retain their privileges after their time-limited certificate expired, approximately one third of the COTH and NACHRI hospitals reported that recertification would need to occur within 1 to 2 years. Paradoxically, 3 hospitals (6%) that said they allowed pediatricians to retain their privileges when their certificate expired reported a contradictory policy that there must be no gap in certification (Table 5). Some hospital personnel stated they could not answer or did not know the answer to the recertification questions because the policy was not formally documented or they had not actually experienced the issue (eg, failure to recertify).
Pediatric Subspecialists. Of the 153 respondent hospitals with subspecialists on their staff, 69 (46%) did require pediatric subspecialists with time-limited certificates to recertify. The COTH- and NACHRI-designated hospitals were more likely to require recertification than non-COTH and non-NACHRI hospitals (Table 3).
Fifty-two (78%) of the 69 hospitals that required pediatric subspecialists with time-limited certificates to recertify allowed pediatricians to retain their privileges even after their certificate expired. Significantly more COTH than non-COTH hospitals (86% vs 60%, weighted P = .01) allowed their subspecialists to retain their privileges.
For those subspecialists allowed to retain their privileges, 33% of the COTH hospitals and 32% of NACHRI hospitals noted that recertification would need to occur within 1 to 2 years or during the next credentialing cycle (Table 5). However, 24% of non-NACHRI hospitals and 23% of non-COTH hospitals reported having no specific time frame for which certification must occur.
Overall, 133 respondent hospitals (85%) stated that they had never revoked or denied privileges due to failure to recertify. A significantly greater percentage of NACHRI hospitals compared with non-NACHRI hospitals reported that they had revoked or denied privileges due to failure to recertify (14% vs 2%, weighted P = .01).
A total of 149 respondent hospitals (94%) did not require permanent certificate holders to recertify. There were no significant differences between the comparative hospital groupings.
Of the 45 nonfreestanding children's hospitals, 39 (87%) reported having certification and recertification policies that were the same as their larger parent institution. The NACHRI nonfreestanding hospitals were more likely to have certification policies that were the same as the larger institution (59% vs 14%, weighted P = .001).
Several of the interviewees reported processes under way to reevaluate the current status of privileging requirements. Although the processes varied greatly among the institutions in their methods and timelines, many expected changes to their rules within the next 2 to 3 years. Several hypothesized that increased use of board certification and recertification was likely.
The most important finding from our study is that although 124 hospitals (78%) did not require board certification at the time of initial privileging of general pediatricians, 111 (70%) did require certification at some point for pediatricians on their medical staff. However, many of these hospitals do not have a specific time limit in which certification must be achieved. Thus, it is possible that the stated requirement for board certification is never actually enforced. Similarly, 113 (74%) of the 153 hospitals studied do not require subspecialists to be board certified at the time of initial hiring; however, only 49 hospitals (44%) required subspecialists to become certified at some (usually unspecified) point in the future. Of the 113 hospitals who did not require certification at initial privileging of subspecialists, 103 (91%) did require these physicians to have at least completed a subspecialty fellowship. For many hospitals, the policy of not requiring certification at the time of initial privileging is likely due to the fact that many new graduates of residency or fellowship may not sit for board examinations for up to 3 years following completion of training.
These results raise a number of issues regarding the manner in which board certification is used or not used by hospitals in their efforts to ensure that high-quality medical care is practiced within their institutions. During the past 2 decades, all members of the ABMS have begun to issue time-limited certificates. The premise for the development of recertification is the need to assure the public of the continued competence of physicians over the course of their professional careers. By embracing the concept of recertification, the boards are making an effort to maintain a level of high standards of medical practice to protect the public by using fair, valid, and reliable methods to assess professional competence.6- 8 However, only 71 hospitals (45%) required general pediatricians with time-limited certificates to recertify, which demonstrates the lack of utilization of this effort and a missed opportunity to provide a measure of quality assurance to patients in their institutions. A total of 48 hospitals (69%) allowed general pediatricians to retain their privileges after a time-limited certificate expired, which further erodes the utility of recertification for patients cared for in US hospitals.
Results for subspecialists were similar with the only exception being that non-NACHRI hospitals and nonteaching hospitals were much less likely to require recertification compared with NACHRI-member institutions and teaching hospitals. This may be due to the paucity of pediatric subspecialists and the need for these hospitals to establish as few barriers as possible to the recruitment of subspecialists to practice at their institutions. As such, market conditions seem to influence the standards by which some hospitals privilege pediatric subspecialists.
The issue of evaluating the currency of time-limited certificates is of growing importance. Because of the changing demographics of the field of pediatrics, 44% of all ABP diplomates already have time-limited certificates. Soon, more than half of board-certified pediatricians will be required to recertify to maintain their status. Hospitals that do not take advantage of this regularly occurring quality assessment in their privileging assessments will be out of step with the self-regulating activities of the pediatric profession.
A recent change to the recertification process for pediatrics has been the institution of secured (proctored) examinations rather than the open-book self-assessment method previously used, which provides the public (eg, families, state licensing boards, hospitals) with a greater certainty that the person completing the examination is the same person completing the recertification process. In addition, this added security has resulted in some states waiving the requirement for a state-administered examination for physicians relocating to that state if an applicant has been recertified by a process that includes a secure examination.4
Despite the lack of recognition of board certification and recertification among many hospitals as a requirement for privileging, the ABMS (the umbrella organization of the specialty boards) is moving forward with an even more rigorous plan to both measure and promote quality of care among physicians of all specialties. The 24 member boards are each developing their own version of a Maintenance of Certification (MOC) program that goes beyond the current requirements for recertification. The components of MOC include evidence of professional standing, lifelong learning and self-assessment, cognitive expertise, and evaluation of performance in practice.9
Our results suggest that specialty boards are attempting to establish higher standards for physicians and patient care than many hospitals in the United States are currently requiring. These efforts by the ABMS are designed to address at least partially the challenges to the medical establishment put forth in the Institute of Medicine reports on patient safety and health care quality.10,11 The ABMS and its member boards believe they can assume a leadership role in reducing the quality gap through MOC.12
Patient expectations regarding the criteria used by hospitals in the privileging process or their assumptions regarding the certification status of physicians providing them with inpatient care is unknown. There have been few studies of patients with regard to their perceptions of the importance of board certification of their physician. Bornstein et al13 studied a sample of more than 600 community residents and found that patients perceived professionally relevant factors, including board certification, as the most important in choosing a physician. Respondents also reported they believed such factors had the greatest effect on the quality of health care they would receive.13 Such findings seem at odds with current hospital privileging policies.
Recently, the American Board of Internal Medicine commissioned a poll among the general public to test its hypothesis that the types of physician quality measures used in the MOC program were of importance to potential patients.5 Their results showed that certification and MOC were highly valued by the public. They also found that patients would change their own physician if they failed to recertify. However, only a minority of respondents ever directly inquired about their physician's certification status.5
In accordance with JCAHO and the National Commission on Quality Assurance standards, professionals must be reevaluated for continuation of privileges at least every 2 years.14,15 However, although certification is required to be assessed at these intervals, it is not mandated to be current by JCAHO.
Regardless of patient preferences, the legitimate question arises as to the veracity of information available demonstrating that board-certified physicians provide care of higher quality than their noncertified counterparts. Some studies have demonstrated better clinical outcomes in hospital settings among certified physicians.16 However, the link between clinical outcomes for patients and the provision of care by a board-certified physician is tenuous, in large part due to the poor quality of many of the studies conducted. A comprehensive review of the literature revealed that only 5% of published studies used research methods appropriate for the research question asked. Of the studies judged to be appropriate, a majority supported an association between board certification status and positive clinical outcomes.17 Certainly, more credible work is required in this area. This is especially important as more certifying boards initiate or continue to evolve existing MOC programs.
The ability of credentialing committees to discern the competence of physicians in the privileging process is difficult and complex, with the role of board certification varying among hospitals. Expectations of the public regarding the qualifications (including board certification) used by hospitals in these efforts also must be considered by those individuals who determine privileging requirements. The establishment of MOC and the institution of recertification by the ABP have created new and potentially useful tools for hospitals to aid in the assessment of physicians. At the same time, efforts must be undertaken by the ABP to assess the impact of recertification and the MOC program on physician quality so that these efforts will provide the maximum utility to hospitals and the patients they serve.
Corresponding Author: Gary L. Freed, MD, MPH, Child Health Evaluation and Research Unit, Division of General Pediatrics, University of Michigan, 300 N Ingalls Bldg 6E08, Ann Arbor, MI 48109-0456 (firstname.lastname@example.org).
Author Contributions: Dr Freed had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Freed, Uren, Hudson, Wheeler, Stockman.
Acquisition of data: Uren, Hudson, Wheeler.
Analysis and interpretation of data: Freed, Uren, Hudson, Lakhani, Wheeler, Stockman.
Drafting of the manuscript: Freed, Uren, Hudson.
Critical revision of the manuscript for important intellectual content: Uren, Hudson, Lakhani, Wheeler, Stockman.
Statistical analysis: Freed, Lakhani, Wheeler.
Obtained funding: Freed, Stockman.
Administrative, technical, or material support: Uren, Hudson, Wheeler, Stockman.
Study supervision: Freed.
Financial Disclosures: None reported.
Funding/Support: This study was supported by the American Board of Pediatrics Foundation.
Role of the Sponsor: The American Board of Pediatrics Foundation did not participate in the design and conduct of the study; in the collection, management, analysis, or interpretation of the data; or in the preparation, review, or approval of the manuscript.