Treatment by a nurse practitioner in primary care improves the severity and impact of urinary incontinence in women. An observational study

Design

We preformed an observational study.

Setting

In this study a total of 16 nurse practitioners, already working with the GP in GP’s
offices, in the eastern part of the Netherlands, undertook a training program in which
they learned how to manage female patients with UI. The training program included
1.5 days course, several home assignments and refresher courses after three weeks,
three and six months. The nurse practitioners were trained in tasks related to diagnostics,
intervention and monitoring of incontinence based on the guideline ‘urinary incontinence’
of the The Dutch College of General Practitioners 5]. All nurse practitioners proved their competence after the course in an assessment.
After the 1,5 days course the participants started to guide patients with UI within
the beginning feedback from the GP and the members of the trainings board till they
have proven their competencies in an individual assessment. At the refresher courses
the nurse practitioners could discuss cases with experts and share their experiences.
The training protocol is available upon request. The GP’s of the practices the nurse
practitioners are working got a 3 h during education course to refresh the guideline
and to explain with expertise they could expect from the nurse practitioners after
the course.

Participants

Women who asked for help in primary care for symptoms of stress-, urge- or mixed UI
were included between June 2009 and December 2010. Patients who underwent an incontinence
operation in de past, patients with a pelvic organ prolapse grade IV (Baden Walker),
a gynecologic malignancy, a neurological disease or unstable mental disorder were
excluded from the study.

Data collection

The GP diagnosed prior to the referral to the nurse practitioner, the type of incontinence
and excluded other causes of urinary incontinence like urinary tract infections or
malignancies. After explanation about the study and informed consent of the patient
to participate in the study the GP gave the patients information about the type of
incontinence but did not explain treatment option. Because co-morbidities and use
of medication can influence UI we also registered them. We compared these data with
the prevalence of the same co-morbidities in a standard primary care population according
to the Continuous Morbidity Registration Nijmegen (CMR) to known if our study population
is a presentation of a general population. The CMR is a very reliable registration
system which determines epidemiological numbers according to the incidence and prevalence
of diseases in primary care for scientific research and education 18]. Because the CMR does not register medication for the aim of research we compared
the use of diuretics and antidepressants with the corresponding diseases such as hypertension/heart
failure and depression as registered in the CMR.

A view days after the consultation with the GP the patients had an appointment with
the nurse practitioner. Prior to the first meeting with the nurse practitioner (T0),
patients filled out questionnaires to assess the severity of the UI (Sandvik-score
and the Patient Global Impression of Severity (PGIS) 19], 20] and its impact on the quality of life (Incontinence Impact Questionnaire) 21]. Patients’ variables like age and variables that may impact UI negatively like mobility
constraints, micturition habits (taking time, have a good position on the toilet,
do not postpone, voiding frequency 5 to 6 times a day), caffeine and alcohol intake
were also mapped. Patients received information and advice and started with conservative
treatment according to the Dutch guideline ‘urinary incontinence’ 5]. After six weeks, the patients returned to the nurse practitioner to evaluate the
effect until then. The patients were motivated by the nurse to continue practicing
and to follow the given advice. After three months (T1), patients filled out the same
questionnaires as they did prior to the first meeting to evaluate the effect of the
treatment (Fig. 1). We chose for a follow up of three months because the aim of the study is to study
if the intervention seems to be effective and to study the feasibility of the intervention.
The dutch guideline ‘Urinary incontinence’ of the The Dutch College of General Practitioners
advise to evaluate the effectiveness of the therapy after three months. If the incontinence
isn’t improved enough a referral to a physiotherapist of urologist/gynecologist can
be considered. If the patient is satisfied with the effectiveness of the therapy it
is important to explain that continuing exercises is needed to sustain this effect.

Fig. 1. Flow chart study population

Outcome measures

Primary outcome

The primary outcome measure was the severity of involuntary urine loss. . The severity
of involuntary loss of urine was measured by the total Sandvik-score (i.e. quantity
of urine loss multiplied by the frequency) 19]. The Sandvik-score of frequency are: never?=?0, 1 time a month?=?1, a few times
a month?=?2, a few times a week?=?3 and every day/night?=?4. The Sandvik-score of
quantity are: none?=?0, drops?=?1, little amount =2 and more?=?3. The total Sandvik-score
can be divided into five categories of severity; none?=?total Sandvik score 0, mild?=?total
score 1–2, moderate?=?total score 3–6, severe?=?total score 8–9 and very severe?=?total
score 12.

Secondary outcome

The secondary outcome measures were the impact of quality of life and patients appraisal
of the severity of incontinence.

The impact on the quality of life was measured by the Incontinence Impact Questionnaire
(IIQ), a questionnaire highly recommended by the International Consultation on Incontinence
(= ICI) and is rated as Grade A 21] The IIQ measures the impact of the UI on daily activities (6 questions), social functioning
(10 questions), emotional wellbeing (8 questions) and travelling/mobility (6 questions).
For each question the patient can choose between: not at all (0 points), mild (1 point),
moderate (2 points) and severe (3 points) 21]. Patients missing 6 or more questions in the IIQ were excluded from further analyses
(N?=?1).

To evaluate women’s overall appraisal of the severity of their UI we asked the patients
to rate their experience of the severity of their incontinence problem with the Patient
Global Impression of Severity (PGIS) into one of four categories (negligible, mild,
moderate, severe) 20].

Sample size calculation

Improvement in the severity of the incontinence was estimated to occur in 40 % of
those in the treatment group compared to 5 % in the group without intervention. Giving
a significance level of 3 % and a power of 80 %, 40 participants were needed. We expected
a drop out during the trial of 30 % and therefore we set our target at 60 patients.

Statistical methods

To test for the changes on the total IIQ-scores between T0 and T1 we used paired t-tests.
To assess the effect of the treatment on the Sandvik subgroups- and PGIS-scores we
used the McNemar test. 22] In order to test for the possible influence of age, the type of incontinence on the
effect measures of the treatment we used a General Linear Model (univariate analyse
of variance) with the dependent variables age and type of incontinence and the independent
variables total Sandvik-. IIQ-and the PGIS-scores.

To investigate differences between the drop-out (patients who started the treatment
but dropped out during the follow up of three months) and the group patients who completed
treatment we used a General Linear Model (univariate analyse of variance) with the
dependent variables age, the total Sandvik- and IIQ-scores and Chi-square tests for
variables type of incontinence, co-morbidity, use of medication and PGIS-scores.

Ethical approval

Upon consultation, the Medical Ethics Committee (CMO region Nijmegen Arnhem) stated
that ethical approval was not necessary because of the non-invasive character of the
study (CMO-nr 2010/460).