Choosing a future from a murky past: A generalization-based model of behavior

Remembering the past appears critical in allowing organisms to detect order in an environment, and hence to behave in accordance with likely future events. Yet the shortcomings of remembering and perceiving typically mean that the remembered past differs from the actual past, and hence that behavior does not perfectly track the structure of the environment. Here, we outline how the process of generalization might be used to understand differences between what an organism does, and the structure of the past and potential structure of the environment. We explore how different sources of generalization - both from within the same stimulus situation, and from different stimulus situations - might be modeled quantitatively, and how predictions made by this modeling approach are supported by research. Finally, we discuss how generalization from multiple stimulus situations, longer-term experience, and from stimulus situations in the past that are not identical to the stimulus situation in the present, might contribute to our understanding of how an organism's experience translates into behavior.

A137 COLONOSCOPY RELATED ADVERSE EVENTS IN A POPULATION-BASED COLON SCREENING PROGRAM

Background

The British Columbia Colon Screening Program (BCCSP) is a population-based program enrolling 50–74 year old individuals for biennial FIT (OC-Sensor, cut-off 10 mcg/g) with follow-up colonoscopy for positive FIT. The neoplasia detection rate is 50–55% and over 75% of colonoscopies have a specimen taken. Previously reported colonoscopy adverse event rates for FIT based screening programs vary widely: 0.03–6.2% and 0–2.7% for bleeding and perforation, respectively. Mortality as a result of colonoscopy is rare but has been reported in 0.0004%-0.0074% of colonoscopies. The rate of colonoscopy related adverse events in BCCSP participants is unknown.

Aims

To determine the rate of colonoscopy related serious adverse events within the BCCSP.

Methods

This is a retrospective cohort study of all participants undergoing colonoscopy in BCCSP from November 15, 2013 to December 31, 2017. BCCSP contacts screening participants by phone 14 days post colonoscopy to determine unplanned medical visits the day prior (during bowel preparation) or following the colonoscopy. Unplanned events underwent chart review if the event was a perforation, cardiovascular or respiratory event, or resulted in death, hospitalization, or significant intervention including repeat colonoscopy, interventional radiology, surgery, blood transfusion, cardioversion, casting of a fracture or suturing of a laceration. Chart review was conducted by a Colonoscopy Lead and reviewed by BCCSP Quality Committee. Unplanned events were defined as serious adverse events (SAE) if they resulted in death, hospitalization or significant intervention and further classified as probably, possibly, or unlikely related to the colonoscopy.

Results

A total of 108,004 colonoscopies were performed. Unplanned events were reported in 1753 participants, of which 586 met criteria for review. Of these, 578 were confirmed unplanned events and 409 were SAEs of which 367 (89.7%) were probably, 22 (5.4%) possibly and 20 (4.9%) unlikely associated with colonoscopy. 36/10,000 colonoscopies were associated with a SAE that was probably or possibly related: perforation in 5/10,000, bleeding 22/10,000. Three deaths occurred in the 14 days following colonoscopy that were probably (2 perforations) or possibly related to the colonoscopy (0.3/10,000).

Conclusions

The BCCSP has a colonoscopy SAE rate in keeping with previous publications, particularly in the context of a very high proportion of procedures associated with polypectomy, a known risk factor for perforation and bleeding. This study will help inform screening participants about the risks of colonoscopy in the BC program. Future studies are required to confirm these rates using hospital admission data.

Funding Agencies

None

Germline truncating mutations in both MSH2 and BRCA2 in a single kindred

There has been interest in the literature in the possible existence of a gene that predisposes to both breast cancer (BC) and colorectal cancer (CRC). We describe the detailed characterisation of one kindred, MON1080, with 10 cases of BC or CRC invasive cancer among 26 first-, second- or third-degree relatives. Linkage analysis suggested that a mutation was present in BRCA2. DNA sequencing from III: 22 (diagnosed with lobular BC) identified a BRCA2 exon 3 542G>T (L105X) mutation. Her sister (III: 25) had BC and endometrial cancer and carries the same mutation. Following immunohistochemical and microsatellite instability studies, mutation analysis by protein truncation test, cDNA sequencing and quantitative real-time PCR revealed a deletion of MSH2 exon 8 in III: 25, confirming her as a double heterozygote for truncating mutations in both BRCA2 and MSH2. The exon 8 deletion was identified as a 14.9 kb deletion occurring between two Alu sequences. The breakpoint lies within a sequence of 45 bp that is identical in both Alu sequences. In this large BC/CRC kindred, MON1080, disease-causing truncating mutations are present in both MSH2 and BRCA2. There appeared to be no increased susceptibility to the development of colorectal tumours in BRCA2 mutation carriers or to the development of breast tumours in MSH2 mutation carriers. Additionally, two double heterozygotes did not appear to have a different phenotype than would be expected from the presence of a mutation in each gene alone.

Payment by salary or fee-for-service. Effect on health care resource use in the last year of life

OBJECTIVE

To investigate the effect of physician payment method on use of health care resources.

DESIGN

Retrospective analysis of patient health care data collected for 3 years (1994 to 1996) from the Vital Statistics Department of the British Columbia Ministry of Health. Billing numbers identified physician payment method.

SETTING

Salaried and fee-for-service primary care practices in the Capital Region District of Victoria, BC.

PARTICIPANTS

A total of 582 patients in their last year of life: 106 were attended by salaried family physicians at a community health clinic; 476 were attended by fee-for-service practitioners. Groups were comparable in age, sex, and geographical location.

MAIN OUTCOME MEASURES

Number and cost of specialist and diagnostic services and medications, number of days in hospital (acute and extended care), and main causes of death.

RESULTS

None of the dependent measures showed any statistically significant differences based on comparisons between many variables for patients in the two groups. Costs of pharmaceutical, specialist, and diagnostic services were not significantly different for the two groups. There were three main causes of death, according to codes on death certificates: heart disease, malignant neoplasms, and cerebrovascular disease.

CONCLUSION

Whether physicians were paid by salary or fee-for-service had no empirical effect on health care resource use.