Machine learning classification of surgical pathology reports and chunk recognition for information extraction noise reduction

Background and aims: Machine learning techniques for the text mining of cancer-related clinical documents have not been sufficiently explored. Here some techniques are presented for the pre-processing of free-text breast cancer pathology reports, with the aim of facilitating the extraction of information relevant to cancer staging.

Materials and methods: The first technique was implemented using the freely available software RapidMiner to classify the reports according to their general layout: ‘semi-structured’ and ‘unstructured’. The second technique was developed using the open source language engineering framework GATE and aimed at the prediction of chunks of the report text containing information pertaining to the cancer morphology, the tumour size, its hormone receptor status and the number of positive nodes. The classifiers were trained and tested respectively on sets of 635 and 163 manually classified or annotated reports, from the Northern Ireland Cancer Registry.

Results: The best result of 99.4% accuracy – which included only one semi-structured report predicted as unstructured – was produced by the layout classifier with the k nearest algorithm, using the binary term occurrence word vector type with stopword filter and pruning. For chunk recognition, the best results were found using the PAUM algorithm with the same parameters for all cases, except for the prediction of chunks containing cancer morphology. For semi-structured reports the performance ranged from 0.97 to 0.94 and from 0.92 to 0.83 in precision and recall, while for unstructured reports performance ranged from 0.91 to 0.64 and from 0.68 to 0.41 in precision and recall. Poor results were found when the classifier was trained on semi-structured reports but tested on unstructured.

Conclusions: These results show that it is possible and beneficial to predict the layout of reports and that the accuracy of prediction of which segments of a report may contain certain information is sensitive to the report layout and the type of information sought.

Statin use and breast cancer survival: a nationwide cohort study in Scotland

Background: Preclinical evidence suggests that statins could delay cancer progression. Previous epidemiological findings have been inconsistent and some have been limited by small sample sizes, as well as certain time-related biases. This study aimed to investigate whether breast cancer patients who were exposed to statins had reduced breast cancer-specific mortality. Methods: We conducted a retrospective cohort study of 15,140 newly diagnosed invasive breast cancer patients diagnosed from 2009 to 2012 within the Scottish Cancer Registry. Dispensed medication usage was obtained from linkages to the Scottish Prescribing Information System and breast cancer-specific deaths were identified from National Records of Scotland Death Records. Using time-dependent Cox regression models, hazard ratios (HR) and 95 % confidence intervals (CI) were calculated for the association between post-diagnostic exposure to statins (including simvastatin) and breast cancer-specific mortality. Adjustments were made for a range of potential confounders including age at diagnosis, year of diagnosis, cancer stage, grade, cancer treatments received, comorbidities, socioeconomic status and use of aspirin. Results: A total of 1,190 breast cancer-specific deaths occurred up to January 2015. Overall, after adjustment for potential confounders, there was no evidence of an association between statin use and breast cancer-specific death (adjusted HR 0.93, 95 % CI 0.77, 1.12). No significant associations were observed in dose–response analyses or in analysis of all-cause mortality. For simvastatin use specifically, a weak non-significant reduction in breast cancer-specific mortality was observed compared to non-users (adjusted HR 0.89, 95 % CI 0.73, 1.08). Statin use before diagnosis was weakly associated with a reduction in breast cancer-specific mortality (adjusted HR 0.85, 95 % CI 0.74, 0.98). Conclusion: Overall, we found little evidence of a protective association between post-diagnostic statin use and cancer-specific mortality in a large nation-wide cohort of breast cancer patients. These findings will help inform the decision whether to conduct randomised controlled trials of statins as an adjuvant treatment in breast cancer.

A realist literature review of transition to adult services by young people with life-limiting conditions in ireland.

Background As a result of improvements in care and treatment more young people with life-limiting conditions are now living beyond childhood, meaning they must make the transition from children's to adult services. The loss of long-standing relationships with providers of children's services combines with poor co-ordination of services to make this a daunting prospect for young people and their families. However, there is little evidence on transition services for young people with life limiting conditions, with few models of good practice in the literature.


Aims The purpose of this review was to determine the factors that promote or hinder the transition to adult services for young adults with life limiting conditions, and identify gaps to be addressed.


Methods A comprehensive search of the literature was undertaken using key terms, of the following databases; MEDLINE and the Cochrane Database of Systematic Reviews. 314 articles were sourced and inclusion and exclusion criteria were applied to highlight the most relevant literature.


Results Studies were reviewed using a realist review approach and three themes emerged from the literature. Barriers and facilitators to the transition process were identified associated with: 1. The patient 2. Parents/carers 3. The organisation.


Conclusion It is unclear from the literature what the specific factors are that promote or hinder the transition process for young adults with life limiting conditions who go through the transition from children's to adult services, therefore, research is required to identify the factors that promote and hinder the transition process in Ireland. This research is currently being carried out by the author as part of Doctoral studies. The three year full time Doctoral study commenced in January 2013 and is funded by the All Ireland Institute of Hospice and Palliative Care.

Fear of oxygen therapy for children in Malawi

BACKGROUND: Although pneumonia is a common cause of death in children in Malawi, healthcare staff frequently encounter patients or carers who refuse oxygen therapy. This qualitative study documents factors that influence acceptance or refusal of oxygen therapy for children in Malawi.

METHODS: Nine group interviews involving 86 participants were held in community and hospital settings in rural and urban Malawi. Eleven in-depth interviews of healthcare staff providing oxygen were held in a central hospital. Thematic analysis of transcripts of the audio recordings was carried out to identify recurring themes.

RESULTS: Similar ideas were identified in the group interviews and in-depth staff interviews. Past experiences of oxygen use (direct and indirect, positive and negative) had a strong influence on views of oxygen. A recurrent theme was fear of oxygen, often due to a perceived association between death and recent oxygen use. Fears were intensified by a lack of familiarity with equipment used to deliver oxygen, distrust of medical staff and concerns about cost of oxygen.

CONCLUSIONS: This study identifies reasons for refusal of oxygen therapy for children in a low-income country. Findings from the study suggest that training of healthcare staff to address fears of parents, and information, education and communication (IEC) approaches that improve public understanding of oxygen and provide positive examples of its use are likely to be helpful in improving uptake of oxygen therapy in Malawi.

'Conditional candour' and 'knowing me': an interpretive description study on patient preferences for physician behaviours during end-of-life communication

OBJECTIVE: To understand patients' preferences for physician behaviours during end-of-life communication.

METHODS: We used interpretive description methods to analyse data from semistructured, one-on-one interviews with patients admitted to general medical wards at three Canadian tertiary care hospitals. Study recruitment took place from October 2012 to August 2013. We used a purposive, maximum variation sampling approach to recruit hospitalised patients aged ≥55 years with a high risk of mortality within 6-12 months, and with different combinations of the following demographic variables: race (Caucasian vs non-Caucasian), gender and diagnosis (cancer vs non-cancer).

RESULTS: A total of 16 participants were recruited, most of whom (69%) were women and 70% had a non-cancer diagnosis. Two major concepts regarding helpful physician behaviour during end-of-life conversations emerged: (1) 'knowing me', which reflects the importance of acknowledging the influence of family roles and life history on values and priorities expressed during end-of-life communication, and (2) 'conditional candour', which describes a process of information exchange that includes an assessment of patients' readiness, being invited to the conversation, and sensitive delivery of information.

CONCLUSIONS: Our findings suggest that patients prefer a nuanced approach to truth telling when having end-of-life discussions with their physician. This may have important implications for clinical practice and end-of-life communication training initiatives.

Report on the 2014 UKONS Annual Conference Belfast, 14–15 November 2014: focus on living with and beyond cancer, patient information and support, and innovations in treatment and care

The UK Oncology Nursing Society’s (UKONS) annual conference focused on three major themes. These were ‘Living With and Beyond Cancer’, ‘Patient Information and Support’, and ‘Innovations in Treatment and Care’. It featured a wide range of presentations, industry satellites, exhibitions, poster discussions. and workshops. Presenters ranged from those eminent in their particular field to those gracing the speaker’s podium for the first time. The rich variety of presentations covered policy, cancer trends, clinical developments, care initiatives, personal development, and advances in practice. There was a strong emphasis on skills, knowledge, values, and attitudes, with the most junior and novice nurses mixing with experienced and highly esteemed practitioners.

The expression of RUNX3 in colorectal cancer is associated with disease stage and patient outcome

RUNX3 is believed to have tumour suppressor properties in several cancer types. Inactivation of RUNX3 has been shown to occur by methylation-induced transcriptional silencing and by mislocalization of the protein to the cytoplasm. The aim of this study was to examine the clinical significance of RUNX3 expression in a large series of colorectal cancers using immunohistochemistry and tissue arrays. With advancing tumour stage, expression of RUNX3 in the nucleus decreased, whereas expression restricted to the cytoplasmic compartment increased. Nuclear RUNX3 expression was associated with significantly better patient survival compared to tumours in which the expression of RUNX3 was restricted to the cytoplasm (P = 0.025). These results support a role for RUNX3 as a tumour suppressor in colorectal cancer.

Utility, Caller, and Patient Profile of a Novel Chemotherapy Telephone Helpline Service Within a Regional Cancer Centre in Northern Ireland

Background: The telephone is an accepted and useful means of communication for the management of patient care. The Chemotherapy Telephone Helpline (CTH) service, located in a large inner-city Trust within the United Kingdom, is a unique nurse-led service within Northern Ireland.

Objective: The objective of the study was to investigate the utility, caller, and patient profile of a novel CTH.

Methods: This was a retrospective study of telephone contacts during 2007 to the CTH. Calls were categorized by caller and patient characteristics, reason for call, and subsequent action.

Results: A total of 7498 calls were made to the CTH during 2007. Of these, 25.6% occurred outside 8AM-4PM. Callers included patients (45.8%), lay carers (31%), and health care professionals (20.5%); 35.2% of calls concerned patients with polysymptomatic problems; 36.8% of calls led directly to patients being medically assessed.

Conclusions: The utility of the CTH service confirms the need of this nurse-led service. This service facilitates access to specialist advice and support for patients, their families, and allied health care professionals.

Implications for Practice: The international significance of these findings for practice includes its demonstration of the multifaceted symptom experience of patients receiving chemotherapy and highlights the importance of rapid access to specialist cancer services for patients and their lay and professional carers. In addition, it demonstrates the capacity of helplines to identify gaps in professional skills and training.

Combined genomic and phenotype screening reveals secretory factor SPINK1 as an invasion and survival factor associated with patient prognosis in breast cancer

Secretory factors that drive cancer progression are attractive immunotherapeutic targets. We used a whole-genome data-mining approach on multiple cohorts of breast tumours annotated for clinical outcomes to discover such factors. We identified Serine protease inhibitor Kazal-type 1 (SPINK1) to be associated with poor survival in estrogen receptor-positive (ER+) cases. Immunohistochemistry showed that SPINK1 was absent in normal breast, present in early and advanced tumours, and its expression correlated with poor survival in ER+ tumours. In ER- cases, the prognostic effect did not reach statistical significance. Forced expression and/or exposure to recombinant SPINK1 induced invasiveness without affecting cell proliferation. However, down-regulation of SPINK1 resulted in cell death. Further, SPINK1 overexpressing cells were resistant to drug-induced apoptosis due to reduced caspase-3 levels and high expression of Bcl2 and phospho-Bcl2 proteins. Intriguingly, these anti-apoptotic effects of SPINK1 were abrogated by mutations of its protease inhibition domain. Thus, SPINK1 affects multiple aggressive properties in breast cancer: survival, invasiveness and chemoresistance. Because SPINK1 effects are abrogated by neutralizing antibodies, we suggest that SPINK1 is a viable potential therapeutic target in breast cancer.

Risks to patient confidentiality when communicating health information to blind and partially sighted patients.

The research reports on a survey of 228 blind and partially sighted persons in 15 health authorities across Scotland. The survey reports data on patient experience of receiving health information in accessible reading formats. Data indicated that about 90% of blind and partially sighted persons did not receive communications from various NHS health departments in a format that they could read by themselves. The implications for patient privacy, confidentiality and wider impact on life and health care are highlighted. The implications for professional ethical medical practice and for public policy are also discussed. Recommendations for improved practice are made.

Patient safety and communication: A new assessment for doctors trained in countries where language differs from that of the host country: Results of a pilot using a domain-based assessment

Objective

Global migration of healthcare workers places responsibility on employers to comply with legal employment rights whilst ensuring patient safety remains the central goal. We describe the pilot of a communication assessment designed for doctors who trained and communicated with patients and colleagues in a different language from that of the host country. It is unique in assessing clinical communication without assessing knowledge.

A 14-station OSCE was developed using a domain-based marking scheme, covering professional communication and English language skills (speaking, listening, reading and writing) in routine, acute and emotionally challenging contexts, with patients, carers and healthcare teams. Candidates (n = 43), non-UK trained volunteers applying to the UK Foundation Programme, were provided with relevant station information prior to the exam.

The criteria for passing the test included achieving the pass score and passing 10 or more of the 14 stations. Of the 43 candidates, nine failed on the station criteria. Two failed the pass score and also the station criteria. The Cronbach's alpha coefficient was 0.866.

This pilot tested ‘proof of concept’ of a new domain-based communication assessment for non-UK trained doctors.

The test would enable employers and regulators to verify communication competence and safety in clinical contexts, independent of clinical knowledge, for doctors who trained in a language different from that of the host country.

A critical ethnography of communication processes involving the management of oral chemotherapeutic agents by patients with a primary diagnosis of colorectal cancer: study protocol

Aim
To describe the protocol used to examine the processes of communication between health professionals, patients and informal carers during the management of oral chemotherapeutic medicines to identify factors that promote or inhibit medicine concordance.

Background
Ideally communication practices about oral medicines should incorporate shared decision-making, two-way dialogue and an equality of role between practitioner and patient. While there is evidence that healthcare professionals are adopting these concordant elements in general practice there are still some patients who have a passive role during consultations. Considering oral chemotherapeutic medications, there is a paucity of research about communication practices which is surprising given the high risk of toxicity associated with chemotherapy.

Design
A critical ethnographic design will be used, incorporating non-participant observations, individual semi-structured and focus-group interviews as several collecting methods.

Methods
Observations will be carried out on the interactions between healthcare professionals (physicians, nurses and pharmacists) and patients in the outpatient departments where prescriptions are explained and supplied and on follow-up consultations where treatment regimens are monitored. Interviews will be conducted with patients and their informal carers. Focus-groups will be carried out with healthcare professionals at the conclusion of the study. These several will be analysed using thematic analysis. This research is funded by the Department for Employment and Learning in Northern Ireland (Awarded February 2012).

Discussion
Dissemination of these findings will contribute to the understanding of issues involved when communicating with people about oral chemotherapy. It is anticipated that findings will inform education, practice and policy.

Patient-reported 'ever had' and 'current' long term physical symptoms following prostate cancer treatments

OBJECTIVE: To document prostate cancer patient reported 'ever experienced' and 'current' prevalence of disease specific physical symptoms stratified by primary treatment received.
PATIENTS: 3,348 prostate cancer survivors 2-15 years post diagnosis.
METHODS: Cross-sectional, postal survey of 6,559 survivors diagnosed 2-15 years ago with primary, invasive PCa (ICD10-C61) identified via national, population based cancer registries in Northern Ireland and Republic of Ireland. Questions included symptoms at diagnosis, primary treatments and physical symptoms (impotence/urinary incontinence/bowel problems/breast changes/loss of libido/hot flashes/fatigue) experienced 'ever' and at questionnaire completion ("current"). Symptom proportions were weighted by age, country and time since diagnosis. Bonferroni corrections were applied for multiple comparisons.
RESULTS: Adjusted response rate 54%; 75% reported at least one 'current' physical symptom ('ever':90%), with 29% reporting at least three. Prevalence varied by treatment; overall 57% reported current impotence; this was highest following radical prostatectomy (RP)76% followed by external beam radiotherapy with concurrent hormone therapy (HT); 64%. Urinary incontinence (overall 'current' 16%) was highest following RP ('current'28%, 'ever'70%). While 42% of brachytherapy patients reported no 'current' symptoms; 43% reported 'current' impotence and 8% 'current' incontinence. 'Current' hot flashes (41%), breast changes (18%) and fatigue (28%) were reported more often by patients on HT.
CONCLUSION: Symptoms following prostate cancer are common, often multiple, persist long-term and vary by treatment. They represent a significant health burden. An estimated 1.6% of men over 45 is a prostate cancer survivor currently experiencing an adverse physical symptom. Recognition and treatment of physical symptoms should be prioritised in patient follow-up. This information should facilitate men and clinicians when deciding about treatment as differences in survival between radical treatments is minimal.

'Talk to me': a mixed methods study on preferred physician behaviours during end-of-life communication from the patient perspective

Background
Despite the recognized importance of end-of-life (EOL) communication between patients and physicians, the extent and quality of such communication is lacking.

Objective
We sought to understand patient perspectives on physician behaviours during EOL communication.

Design
In this mixed methods study, we conducted quantitative and qualitative strands and then merged data sets during a mixed methods analysis phase. In the quantitative strand, we used the quality of communication tool (QOC) to measure physician behaviours that predict global rating of satisfaction in EOL communication skills, while in the qualitative strand we conducted semi-structured interviews. During the mixed methods analysis, we compared and contrasted qualitative and quantitative data.

Setting and Participants
Seriously ill inpatients at three tertiary care hospitals in Canada.

Results
We found convergence between qualitative and quantitative strands: patients desire candid information from their physician and a sense of familiarity. The quantitative results (n = 132) suggest a paucity of certain EOL communication behaviours in this seriously ill population with a limited prognosis. The qualitative findings (n = 16) suggest that at times, physicians did not engage in EOL communication despite patient readiness, while sometimes this may represent an appropriate deferral after assessment of a patient's lack of readiness.

Conclusions
Avoidance of certain EOL topics may not always be a failure if it is a result of an assessment of lack of patient readiness. This has implications for future tool development: a measure could be built in to assess whether physician behaviours align with patient readiness.