Health Secretary Andrew Lansley’s pledge to end mixed sex accommodation in the NHS made quite a splash amongst patients' groups and media outlets.

 

Many headlines noted that current single sex regulations were breached some 64,000 times. Looking at this figure however, Full Fact found some grounds for caution.

 

The number originates from the Department of Health press release which in turn references a survey of Strategic Health Authorities (SHAs) conducted on its behalf.

 

What doesn’t make it into the newspaper reports however are the caveats attached to this study. Of the 152 SHAs in England, only 74 returned data. A further 68 had no relevant records, and 8 SHAs had not returned the survey by the deadline.

 

Of the 74 positive respondents, five did not categorise the breaches into ‘clinically justified’ and ‘non-clinically justified’ as required by the survey, meaning that these were also removed from the final results.

 

From the 69 SHAs with complete data, the Department of Health was able to conclude that there were 8,028 breaches over the first quarter of 2010/11. This number was doubled to account for the SHAs not returning data, and multiplied by four to give an annual figure of 64,224.

 

There are a number of reasons why this calculation may be a little crude.

 

Firstly, extrapolating annual trends from a single quarter’s worth of data brings with it a few problems. According to the previous Government’s Delivering Same-Sex Accommodation (DSSA) initiative, breaches of single sex regulations occurred more frequently when staff and resources were stretched.

 

As hospital inpatient admissions can vary with the seasons (for example, in response to a heatwave or heavy snowfall), non-seasonally adjusted figures for any given quarter may not accurately reflect annual trends.

 

As Department of Health Hospital Episode Statistics show, certain months, such as March and October, consistently have higher inpatient admissions than others.

 

Furthermore, the fact that the identities of the SHAs that responded to the survey were withheld means that it is difficult to gauge the weight that should be attributed to the positive data that was returned.

 

If, for instance, DSSA reporting was more widely adhered to in larger SHAs than in smaller authorities, it would be reasonable to assume that the number of breaches not included in the survey would be smaller than those which were. This would cast doubt on whether the Department’s multiplication of the 8,028 by two gives an accurate reflection of the true situation.

 

Full Fact contacted the Department of Health on this point, and they were unable to give any indication of the size of type of SHA that returned positive data as part of the survey.

 

However they were able to confirm to us that the survey was “informal” and does not correlate with any historic official data. It is therefore also difficult to place it in context, or measure the progress that has been made in this area.

 

So whilst there is no reason for us to think that the Department’s calculation of the 64,000 figure is anything other than the best estimate possible from a limited data set, these limitations mean that caution must be applied before any great significance is attached to them.