GI endoscopy in rural communities: Experience of a rural family physician in British Columbia

Background
Digestive disorders are extremely common in the general population, especially in the elderly. Endoscopic examination is important in evaluating and managing upper and lower gastrointestinal problems. About 1% of the general population has an endoscopy each year, although this proportion varies widely across health districts and regions.[1,2] 

An endoscopic diagnosis is generally agreed to be required before long-term treatment is initiated. Evidence is accumulating that early endoscopy is more cost-effective than empirical treatment followed by endoscopy when indicated.[3] 
As well as playing a role in eval­uating problems, endoscopy plays a role in screening for colon cancer. There is a national movement to screen patients at risk for colon cancer, and a growing need for surveillance of patients with a history of polyps and colorectal cancer. 

Many physicians and patients prefer colono­scopy as the reference standard screening study for colon polyps and cancer as well as for surveillance.[4] The lack of qualified endoscopists in rural areas means that family physicians with training in upper and lower gastrointestinal endoscopy are often required to perform these procedures. 

The issue of how we define a “well-trained” family physician endoscopist is a  matter of controversy. There is no consensus and no national or provincial certification process that is required for endoscopic privileges in many Canadian hospitals. 

However, most leaders in training endoscopists em­phasize that clinical knowledge and judgment are at least as important as technical skill in manipulating a scope, and recommend strongly that all  programs for certifying endoscopists incorporate such clinical training. 

In order to determine the quality of endoscopic examinations performed by a rural family physician, a study was initiated to collect some baseline data.

Methods
A literature review was conducted using the following keywords and subject headings: endoscopy, digestive system, gastrointestinal, rural health, rural hospitals, general practitioner, family physicians, and Canada. Only two articles specifically about GI endoscopy in rural communities were found.[5,6] 

One study looked at  the safety and effectiveness of colono­scopy done by family physicians in rural Ontario.[5] The other study looked at patterns of upper and lower GI endoscopy in Alberta.[6] While there are very few studies addressing the provision of GI endoscopic services in small rural hospitals by nonspecialist physicians, there are numerous studies about the quality and safety of the procedure in general. 

These studies report that a competent colonoscopist can reach the cecum more than 90% of the time.[7-13] (There is a lack of data on reaching the duodenum, even though this might prove to be an indicator of the colonoscopist’s competence.) 

Studies also show that endo­scopy can be completed with a complications rate of 0% to 5%,[8,10-15] and with a satisfactory rate of detection of pathologic lesions.[8,12-16] The complications reported by various research­ers include arrhythmia, bradycardia, hypotension, hypoxia, bleeding, and perforation.

The retrospective observational study that followed the literature review used data collected on 100  pa­tients undergoing endoscopic exam­inations performed by the same rural family physician at St. John’s Hos­pital in Vanderhoof, BC. 

This 24- bed health care facility in central Bri­tish Columbia serves a population of 12 000 people. From September 2002 to December 2004 (2.5 years), 677 scopes were undertaken at the facility. One hundred charts describing 49 upper and 58 lower scopes (some pa­tients had both) were randomly selected and every third chart was analyzed in detail. 

The data collected included patient demographic characteristics, indications for endoscopy, endoscopic and pathologic findings, and complications. Analysis of these data provided some quality indicators: the reach-the-cecum-rate (RCR), the reach-the-duodenum rate (RDR), the complications rate, and the false-negative and false-positive rates.

Results 
Of the 100 patients whose charts were reviewed, 49 were female and 51 were male. The mean age was 53.1 years. The most common indication for an upper scope was chronic abdominal pain (38.8%), and for a lower scope was PR bleeding (36.2%). Only one colon cancer and one esophageal cancer were found (Table). 

The phy­sician completed the procedure in all but 4 cases, and reached the cecum (56/58) or duodenum (47/49) as requir­ed in almost all cases. There were no complications, 6 false-negative results, and 11 false-positive results. Overall these results compare favorably with those expected of a competent endoscopist.

According to the American Society for Gastrointestinal Endoscopy (ASGE) and the Canadian Association of Gastroenterology (CAG), the following are some of the appropriate quality indicators when reviewing GI endoscopy for both clinical and re­search purposes: sedation, complications, patient satisfaction, duration of procedure, and success of procedure (technical success, subjective outcome, and diagnostic findings). 

Many studies rely on the modified GHAA-9 patient satisfaction survey, a useful tool that has been validated in numerous patient populations and has been used for more than 20 years (Figure). Technical success is commonly de­fined as reaching the cecum in lower endoscopy and reaching the duodenum in upper endoscopy. 

Other indicators of procedural success are subjective outcome (whether the findings were clinically useful), and diagnostic findings (visual and/or pathologic). Time taken for colonoscopy is not itself a measure of quality, but is of interest when considering efficiency and scheduling. The average time for colonoscopy in different studies varies.[15] 

However, it is now recognized that for polyp screening, the operator should take a minimum of 6 minutes to withdraw the scope from the cecum, and that using less time may result in a higher miss rate for small neoplasms. In larger centres, taking longer than 15 minutes on average for insertion and withdrawal results in lack of efficiency and further prolongs waiting lists. 

It would be difficult to measure whether this need for efficiency exists to the same degree in rural communities, or in fact what the relative per capita cost is for endoscopy in different communities. There is a lack of reported time studies for upper GI endoscopy.  

Limitations of the study discussed here include a relatively small sample size and the fact that only one rural family physician performed all the procedures. Data on certain suggested criteria were not collected (e.g., patient satisfaction and duration of procedures). 

In addition, no pathologic report was available for comparison in 30% of cases, and 4% of procedures were incomplete. Future studies are needed to specifically address these limitations and other issues. While the measures studied assessed the endoscopist’s accuracy in recognizing mucosal pathology, the real value of endoscopic diagnosis is in confirming or refuting a clinical diagnosis, not in verifying gross mucosal appearance by means of a biopsy. 

The endoscopic diagnosis cannot be the main determinant of real outcome unless des­cribed in the context of detailed history and physical exam. For the same reasons, the clinical management of the presenting problem during the procedure was not described in this study.  For example, were polyps removed at colonoscopy once identified? Were bleeding lesions treated by appropriate endoscopic means? 

These would be useful data in assessing the overall effectiveness, efficiency, cost-effectiveness, and productivity of the procedure, as well as designing specific training programs and courses for potential candidates.

One strength of this study is the centrality of information: with only one hospital and medical clinic in town, the chance that any complication would be missed is unlikely. Another potential strength is the use of RDR data, which might be utilized for further comparative studies in the future. 

Conclusions 
In a central BC hospital, a well-trained rural family physician was able to provide safe and technically competent diagnostic and therapeutic GI endos­copy for patients. 

The quality of the physician’s results compared favorably with published results for the procedures performed. However, we still need more information on what type of procedural instruction and what amount of experience is necessary for family physicians performing endos­copy in rural areas. 

Furthermore, outcome studies are needed to relate endoscopic findings to clinical diagnosis and patient management, and to determine how these relationships are affected by the practitioner’s training.  

Further prospective research using recommended quality indicators will assist in establishing objectives and standards for training programs for rural practitioners and will allow patients in small communities to benefit from the presence of physicians who are well trained in this area of practice. 

Competing interests
None declared.

Acknowledgments 
I’d like to thank Dr Sean Ebert for performing the procedures and providing useful feedback, and Dr Jonathan Berkowitz for assisting with statistical analysis of the data. Thank you to Ms Areta Wong for her assistance in formatting the manuscript.

Dr Vala is a family physician in North Vancouver, BC. He has practised in a variety of rural and urban parts of Iran and Canada.