Dear Dr Goldbeck-Wood,
Thank you for accepting our work for publication. We have revised the paper to meet the requirements of the short report format. We have addressed each point in the report from the BMJ’s full Editorial Committee, and the checklist. Enclosed is the revised paper as hard copy and on floppy disk, as requested.
| 1. | The SF36 details have been removed from all sections of the paper |
| 2. | We have moved the explanation of pension enhancements from the body of the text to an Annex (web only), and expanded it for clarity. |
| 3. | We have updated our literature search and could find no reference to ill-health retirement rates in a named industry in the UK, or any industry abroad. The only relevant reference remains that which we have used. However this does describe ill-health retirement rates for 6 UK industries (four public and two private sector). We have therefore replaced our previous statement with one describing the range of rates for the 6 industries as this should be of greater interest to an international readership. |
| 4. | The conclusion that "a third of cases may be preventable" has been omitted. |
| 5. | We based our comment about ‘losing highly experienced employees’ on our finding that retirees had a mean of 21.2 years of NHS service. This was reported in the results section along with the range. We have added to the text "87 % had worked for at least 10 years in the NHS". We believe it remains useful to report both the mean and range, and hope that you agree. |
| 6. | We can reassure you that ethics committee approval was obtained for this study, and prior to the start of data collection, from the Royal Free Hospital and Medical School ethics committee. For retirees who did not consent to take part in the prospective questionnaire study, the NHS Pensions Agency provided anonymised data which is routinely published by them including age, sex, job title and medical reason for ill-health retirement. |
| 7. | From our experience of working in the NHS we have no reason to believe that there is a seasonal variation in ill-health retirement applications. This was confirmed by discussion with the NHS Pensions Agency at the time we designed the study. We therefore chose to use consecutive applicants, so that we could collect the data over a shorter time period (6 months). This should also make any prospective data collection intervals more contained. |
| 8. | We considered carefully the sample size at the design stage of the study. A power calculation was not appropriate and as there is very little literature in this area, we chose to study the largest sample size we believed we could manage and which the NHS Pensions Agency would agree to assist us with. |
Checklist
We have added study design (cross-sectional) to the study title.
We have included email address of author for correspondence
We have moved the competing interests statement to before the references
We have spelt out the abbreviations you listed.
We have reduced the paper to 621 words (excluding title), one table and five references
We have italicised the titles of the journals and other documents referenced
We have laid out the title page in keeping with the guidance and followed the structure for a short report
We believe the above answers all of your queries and requests. Do let us know if there is anything further we need to do,
Best wishes,
Shriti Pattani, Nick Constantinovici, Siân Williams