{
  "abstract": "Background Limited data exists on cancer outcomes for LGBTQ+ individuals and those with English as a second language. Evidence indicates these groups may experience multiple barriers to cancer care and it has been hypothesised this may lead to later diagnoses, poorer prognoses and higher mortality. Official cancer statistics in the UK do not currently include sexual orientation or language status as a standard demographic breakdown. However, this information is included in England’s national Cancer Patient Experience Surveys (NCPES), so is available for survey respondents.Methods Through an ongoing analytical partnership between NHS England’s National Disease Registration Service and Macmillan Cancer Support, NCPES data from 2017–2022 has been linked at the patient level to cancer registration data using a multi-stage linkage process based on NHS numbers and tumour site.Multinomial logistic regression models were used to analyse associations between people’s sexual orientation (categorised as either heterosexual or LGB+), English language status and their cancer outcomes (including stage at diagnosis and survival), adjusting for confounders including age, deprivation, and tumour type. Non-respondents to the relevant questions in NCPES were excluded from this analysis.Results The dataset included roughly 51,000 patients annually (approximately 16% of all cancer diagnoses), with around 1.5% (730) LGB+ respondents and 4.5% (2,400) with English as a second language each year. Survival rates for LGB+ respondents were 98.8% at 1-year and 80.5% at 5-years post-diagnosis. For those with English as a second language, survival rates were 99.1% at 1-year and 83.0% at 5-years. Stage 3 or 4 diagnoses were 42.2% for LGB+ respondents and 41.2% for those with English as a second language.Following log rank tests and regression modelling we did not identify any statistically significant differences in survival for the two groups. Those with English as a second language had slightly lower odds of late-stage diagnosis (p<0.05). However, power calculations indicated insufficient sample sizes to draw definitive conclusions.Conclusion This novel analysis provides estimates using national datasets of cancer survival and stage at diagnosis for cancer patients in England who are LGB+ or who have English as a second language. These estimates are biased by the cohort of the survey (those with the poorest prognosis will not be alive or able to complete NCPES, respondents must have had inpatient cancer care to be eligible for the survey, those less able to read and write English may not have been able to complete the survey and survey respondents tend to be younger and less deprived).",
  "authors": [
    {
      "affiliations": [
        "Insight and Performance, Macmillan Cancer Support, London, UK"
      ],
      "name": "Lizzie Augarde"
    },
    {
      "affiliations": [
        "Insight and Performance, Macmillan Cancer Support, London, UK"
      ],
      "name": "Luke Morton-Holtham"
    },
    {
      "affiliations": [
        "Insight and Performance, Macmillan Cancer Support, London, UK"
      ],
      "name": "Rachel White"
    },
    {
      "affiliations": [
        "Insight and Performance, Macmillan Cancer Support, London, UK"
      ],
      "name": "Rachel Bowden"
    }
  ],
  "title": "P12 Clinical outcomes for cancer patients in England by sexual orientation and language status: novel analysis of linked national cancer patient experience survey data",
  "uid": "b3bf542c-6b85-5a79-9623-1aa50b9fd83a"
}