Pressure ulcers are a common phenomenon in many different health care settings. Over the years several Dutch studies show indeed that the prevalence of pressure ulcers is high, especially in the intensive care unit (Bours, 2001). There is a need for an effective therapy of managing these ulcers. Pressure ulcers are medically known as decubitus ulcers. A decubitus ulcer is a pressure sore or what is commonly called a bed sore. It can range from a very mild pink coloration of the skin, which disappears in a few hours after pressure is relieved on the area, to a very deep wound extending to and sometimes through a bone into internal organs. There has been a trend in modern health care toward minimally invasive procedures, including reduced reliance on heroic and long term drug therapies (Kenneth, 2006). The trend for managing pressure ulcers has been towards the use of micro-current. living tissue possess direct current electro-potentials that appear to regulate, at least partly, the healing process. Micro-current therapy is the practice of applying low-intensity electric currents, usually at low frequencies that match the bodys natural pulse rate. Micro-currents, or micro-amps, are electric currents with intensity less than one milliamp. Micro-currents are measured in the millionth of an amp range. However, the evidence that micro-current therapy predictably accelerates dermal (skin) repair remains less convincing (Mohammad, 2006). In view of recent scientific understanding of the wound-healing process, one would expect a beneficial outcome from electro-therapy that decreases ulcer size and accelerates healing in patients (Gentzkow, 1993). Chronic wounds, of which leg ulcerations make up a major share, are a therapeutic problem. It is estimated that 90% of leg ulcers are due to venous stasis, affecting 0.6 of men and 2.1% of women in their 60s (Stiller and Dermato, 1992; Nessler and Mass, 1985).
A statistical analysis of data and results from a prospective study, undertaken with the aim to detect the effect of Micro-current Electrical Therapy (MET) on patients hospitalized for a long time and therefore baring decubitus wounds.
MATERIALS AND METHODS
The data was collected from patients hospitalised for a long time and therefore
suffering from decubitus wounds in Belgium. A total of 60 male and female patients,
aged between 60 and 80, got enrolled into the study. These were from 6 different
hospitals and got randomised into 2 different groups. The randomization was
important so that bias is removed and the two groups are comparable.
|| The variable description, type and units of measure
|*CON meant that the patient received micro-current therapy
but from a device which was not working properly and therefore belonged
to the control group, while MET meant that the patient received micro-current
therapy from a properly working device and thus belonged to the treatment
or case group
Patients from one group, MET group, were receiving micro-electronic therapy
while those in the other group, CON group, were receiving visually the same
therapy but with a micro-electronic device which was not working properly. This
second group served as the control group. All patients were followed for 12
weeks and the surfaces of the wounds measured on a weekly basis. Some patients
had more than one wound.
The variables were collected and made up the data set (Table 1). It also bares the names given to these variables in the data set and information on their units of measure.
Derivation of the response variable: Our data set contains 114 subjects; wounds. The measurement of surface area for each wound was taken for 12 consecutive weeks. We denote the measurements per a wound as shown below Y0, Y1, Y2,..., Y12 where Y0 is the base measurement of the wound and t = 0,1,2,...,12 weeks. Assuming that the surface of the wound at time t is a function of the surface at time t-1, we can derive the rate of change in surface area of a wound for the 12 weeks as follows.
Let Y1 CYt-1. This implies that, Y1 = CY0, CY2, = CY1, Y3 = CY2,..., CY12 = CY11 where C is unknown constant. Using mathematical equation, we come up with an equation that tracks the correlation between all 13 measurements.
To solve for C, which is our main interest, we applied logarithms on Eq. 1
Combining Eq. 2 and 3, we get,
Equation 4 can be interpreted as a linear regression model
without intercept, that is, Zt = (in C)* t + ε. There was no
change in surface area of a wound at week 0. We regressed Zt on t
for each wound. The parameter estimate,
from the regression model is equivalent to ln C. That is, ln C =
⇒ C = exp ()
Our response variable C is the rate of change in surface area of a wound, where
||C = 1
|| corresponds to no change in the surface area
||corresponds to improvement; the surface area is reducing
||corresponds to deterioration; the surface area is increasing
Regression model: To examine the treatment effect, multiple linear regression models were used. Cohen and Hardy proposed that any combination of categorical and continuous variables can be analyzed within a multiple regression model framework simply through the dummy coding of the categorical variables (Cohen, 1968). In this study we, therefore, chose to assess the treatment effect using a multiple linear regression model (Kutner et al., 2005).
Therefore, in present study:
|| Rate of change in surface area of wound (C)
|X4, X5, X6,
|| Hospital (H) 1, H2, H3, H4 and H5, respectively
RESULTS AND DISCUSSION
Overall, healing rate ranged from 0.7340 to 1.0838, with mean 0.9240 and standard deviation 0.0807. In the CON group, the healing rate ranged from 0.8551 to 1.0838, with mean 0.9455 and standard deviation 0.076, while it ranged from 0.7340 to 1.0511 in the MET group, with mean 0.9455 and standard deviation 0.0760 (Table 2).
Overall, the age ranged from 60 to 79 years in the CON group, with mean 69.33 and standard deviation 6.23. In the CON group, age ranged from 60 to 79 years, with mean 69.6 and standard deviation 5.93, while it ranged from 61 to 78 in the MET group, with mean 69.07 and standard deviation 6.72 ( Table 2).
In general, 40.35% of wounds were from female and 59.65% from male patients. In the CON group the distribution of wounds by sex was about 23.33% from male and 76.76% from female, while in the MET group it was about 59.30% from male and 40.70% from female patients (Table 3).
Overall patients hospital 2 had the most wounds (33%). Most of the wounds in the CON group were at hospital 2 (43%), while most of the wounds in the MET group were at hospital 6 (30%) ( Table 4).
About two-thirds (67%) of the patients had more than one wound on their bodies (Table 5).
It can be seen that the surface-base ratio is generally decreasing in both groups, but is decreasing at a faster rate in the MET group (Fig. 1), suggesting that healing rate in this group is better than in control group.
The wounds for patients in hospital 4 and 5 are reducing in surface-base ratio faster than other hospitals (Fig. 2). There is no much difference in the surface change of wounds for patients in Hospital 1, 2, 3 and 6. This led us to conclude that the healing rate was different among the hospitals.
Treatments are significantly related with sex and slightly related with hospital. Sex and hospital also significantly associated (Table 6).
The model explained approximately 46% of the total variation by the repressors
(Table 7). Individually every variable has significant effect
on healing rate except hospital 1. Here we also seen that average rate of change
in surface area of wound decreases approximately 3% more using micro-current
electrical therapy than control when all other variables hold constant. That
is Micro-current Electrical Therapy (MET) significantly contribute for reducing
the rate of change in surface area of wound.
|| Summary statistics for the rate of change of surface area
of wound and age
||Frequency distributions of sex in the control and the MET
|| Frequency distributions of sex and hospital in the control
and the MET groups
|| Frequency of wounds per patient
||Multiple linear regression analysis of rate of change in surface
area of wound
|R2 = 0.462, F8,105 <0.001
The average rate of change in surface area of wound reducing approximately
5% more of female patients than that of male patients keeping other variables
||Weekly surface-base ratio for the control and MET groups (Comparison
of MET and control group)
||Comparison of weekly surface base ratios at the hospitals
Similarly among hospitals, the average rate of change in surface area of wound
decrease more (about 17%) those patients who are in hospitalized in hospital
4 than that of patients who are hospitalized in hospital 6 keeping other variables