24 December 2010
Marlyn Glen has been informed by NHS Tayside that the board did not apply statistical tests to health data on Lochee and Dundee to determine if there were any significant differences between them.
( see the text of the letter below)
The data was presented to a recent meeting of the Lochee Community Planning Partnership
The board said that it was “not common practice to apply statistical testing to the local profile data”.
Ms. Glen said that “great care” had to be taken in dealing with data for small areas, particularly in the use of percentage increases in health indicators instead of the straightforward numerical increase.
In one set of data it was stated that there was a 260 per cent rise in the drug- related admission rate to hospital from Lochee over 2 years.
In reply to Ms. Glen, the board said that “260 per cent” increase came from a rise in admissions from 16 to 46 over two years.
The health board revealed to Ms. Glen that on several health indicators Lochee was amongst the highest” for Local Community Planning Partnerships in Dundee “but not the highest”.
These were for :
Estimated prescriptions for anxiety, depression or psychosis
Emergency admission rate
Admissions for alcohol or drugs misuse
Mothers smoking at antenatal booking
First time mothers aged 19 and under
Ms. Glen commented,
” I welcome the clarification given by NHS Tayside, and they are to be commended on their use of such data for small areas that is absolutely necessary for detecting and following health trends and then intervening accordingly.
“However, for greater public awareness and involvement, it might have been better if health data for Lochee that was expressed as rates ‘per 100,000 population’ was given as straightforward numbers, particularly when the population of Lochee was taken to be just over 18,000.
“Furthermore, it has to be remembered that the ‘Lochee’ portrayed in these figures has boundaries that stretch well beyond the realm of traditional Lochee.”
Text of letter from NHS Tayside to Marlyn Glen
It is well documented that within Dundee City there is evidence of health inequalities, the differences in health experiences and outcomes, across areas; in particular, a greater burden of disease and poorer outcomes are demonstrated in areas of deprivation.
With this in mind, a health profile was produced for each of the local community planning partnerships areas with the intention of providing information for local community workers on the data relating to the health inequalities in their particular areas.
The profiles were intended to be used to help target particular pieces of work in the local community planning partnerships areas.
Individual profiles are of local interest, but if taken out of context, may paint a rather gloomy picture and may exaggerate the position of a particular area, when compared to the average situation rather than viewed in conjunction with the other local community planning partnerships areas.
Numbers are small in some cases and any increase in small numbers will produce a potentially alarming increase in rate.
For that reason, it is essential that trends are viewed over a reasonable period to establish what may be random fluctuations.
The explanations and further comments on the tables as requested follow:
It is not common practice to apply statistical testing to the local profile data.
Rather, our aim is to establish if there are gaps across areas and to identify which areas suffer the greatest inequalities.
Table 15: Comparative illness rate per 100,000 population
The range of rates was 9,536 – 20,118; Lochee (17,540) was among the highest but not the highest
Table 18: Estimated prescriptions for anxiety, depression or psychosis
The range of rates was 10.5% – 13.5%; Lochee (13.3%) was among the highest but not the highest
Table 19: Emergency admission rate per 100,000 population
The range of rates was 7,788 – 14,600; Lochee (13,472) was among the highest but not the highest
Table 20: Elective admission rate per 100,000 population
The range of rates was 6,451 – 10,759; Lochee (9,031) was in the middle of the range
Table 21: All cause mortality rate per 100,000 population (<75 years)
The range of rates was 317 – 734; Lochee (641) was among the highest but not the highest
Table 22: Admissions for alcohol or drugs misuse
Alcohol: The range of rates was 514 – 1,348 ; Lochee (1,243) was among the highest but not the highest
Drugs: The range of rates was 0 – 306 ; Lochee (306) was the highest
In Lochee, there was an increase in drug related hospital admissions from 07/08 – 08/09 but the previous two years had remained stable. Other areas had also increased over the period but more gradually. Numbers here are small (admissions increased from 16 to 46) and although this was a substantial increase over a short period, longer term data may show this to be a random peak. Numbers may refer to repeat admissions of individuals.
Table 23: Mothers smoking at antenatal booking
The range of rates was 6.7% – 32.7% ; Lochee (29%) was among the highest but not the highest
Table 24: First time mothers aged 19 and under
The range of rates was 4.3% – 25.4% ; Lochee (20.8%) was among the highest but not the highest
Filed under: Uncategorized