[Qualitative research methods in clinical practice]. / Kvalitativa forskningsmetoder i klinisk medicinsk praxis.

Evidence-based medicine (EBM) corresponds to the physician's ambition to integrate his/her experience with the best current scientific knowledge, applying this integrated proficiency to clinical problems. The concept of EBM has become widely accepted and given rise to a global network usually emphasizing the randomized clinical trial (RCT) as the most important scientific method. However, clinical situations are very complex. To a greater or less extent they are all composed of diverse factors, ranging from bio-medical ones to inter-subjective relations, values and beliefs. In addition to RCT and other quantitative procedures, qualitative scientific methods could be used to explore complex clinical problems. Clinical problems always deserve to be addressed from various perspectives. Unfortunately, RCT has acquired such a dominant position as to give the impression that it is the sole preferred method within the EBM framework. Based on three examples, the present article claims that it is the questions asked by an investigation that should determine one's choice of methods, not the reverse. Consequently, we argue that it is important to remember that RCT is just one out of several potential scientific methods. The respective utility of any one method depends on the clinical issue at hand.

Short or long consultations? Primary care nurses' considerations in making appointments with general practitioners.

In Sweden primary care nurses (PCNs) at the health centres (HCs) manage the appointment systems and decide the appointed length of general practice consultation time. This study aims at elucidating experienced PCNs' considerations in booking consultations of different lengths. The grounds for giving long or short appointments were searched for. Qualitative methods were used and 10 PCNs at eight HCs in Umeå Health District in northern Sweden were interviewed. The PCNs take factors from two areas into consideration; one area, often immediately apparent, concerns immediate and straightforward issues such as the patient's reported problem, the patient her/himself, the doctor her/himself and the current situation at the HC and in the catchment area. The other area for consideration usually develops through experience over the years and concerns time consumption in a longer perspective. The overall quality of the consultations is then in focus and the length of a separate consultation is of minor interest. The PCNs do not applicate a straightforward relation between length of time and quality of consultations. Experienced PCNs seem to promote accessibility of primary care and their efforts in allocating time in a practical way probably increase the quality of general practice consultation.

Separation and distress--sickness absence and health screening in newly divorced middle-aged Swedish men.

The present study aimed at a description of the sickness absence pattern during 1982-1989 in 32 men who divorced in 1984. Another purpose was to carry through a health screening of the men within six months after the marital disruption focusing on risk factors concerning cardio-vascular disease (smoking, overweight, hypertension) and high alcohol consumption (elevated GGT). The year of divorce and the successive three years (1984-1987) were characterized by high sickness rates (average 21,7 days/year, variation 19,4-26,6) compared to a reference group (average 16,6, variation 14,9-18,1). In the remaining four years (1982-1983 and 1988-1989) the sickness absence was lower in the divorced group (average 12,2, variation 8,7-18,0, reference group: average 17,3, variation 14,8-20,0). The increase was mainly due to short absence periods (self-certifications). The health screening (health examination and record analysis) (n = 29) revealed high frequency of daily smoking and alcohol overconsumption. Overweight and hypertension were not overrepresented. The findings are discussed in relation to a supposed male reaction style to separation. The impact of social isolation is stressed.

Life after divorce--a five-year follow-up study of divorced middle-aged men in Sweden.

Twenty-seven men, between 35 and 49 years of age, were asked via a questionnaire for information about their life circumstances (housing and working conditions, economic situation), actual symptoms, perceived health and health-related behaviour 5 years after they had divorced. The information was compared to corresponding data from an earlier study carried out within 7 months after the legal divorce. Housing conditions and personal economy had improved and working conditions were still satisfactory. Perceptions of health were more positive and in particular earlier, frequently recorded psychological symptoms (uneasiness, anxiety, depression) had diminished. Alcohol and smoking habits remained unchanged and common, especially among manual workers with more than one divorce. Men living in a stable relationship with a new partner or who had custody of children had fewer social and medical problems than men living alone. These findings support the connection between social support and health.

Premature deaths among men in a Swedish municipality--social characteristics prior to death.

Premature deaths among 91 men in a Swedish municipality were analyzed regarding social class, immigrant status and their contacts with the Social Services because of alcohol abuse. Unskilled manual workers were overrepresented (54% versus 28% in an age-adjusted male reference group) and self-employed (other than professionals) were underrepresented (4% versus 15%). The job situation was, however, rather stable and low unemployment (1%) was at hand. The proportion of men born in Finland was high (20%). All men in this group belonged to the lowest social class. Single men (unmarried or divorced) were overrepresented compared to the Swedish group (72% versus 40%). Cardio-vascular disease was a common cause of death among the Finnish men (56% versus 36%). Combining medical and social data revealed alcohol abuse among 40% of the 91 men (50% among the Finnish and 37% among the Swedish men). The most complete information was given by the primary health care records where 89% with alcohol abuse were identified, which is a high proportion compared to other studies. Only 28% were considered to have alcohol abuse according to the Social Services.

Premature deaths among men in a Swedish municipality--civil status and primary health care utilization prior to death.

Causes of premature deaths among men in relation to civil status and primary health care utilization were studied in a Swedish municipality. Alcoholism, suicide and accidents (psycho-social deaths and accidents) were of special concern (40% of all deaths). When principal and contributing causes of death were considered, alcoholism was registered in 21% of all men. Including information from primary health care records, the Social Services and the local outpatient alcohol clinic, the proportion of men with alcohol abuse was 40%. Half of the men with psychosocial deaths and accidents were divorced at the time of death. The time elapse between divorce and death was especially short for those who committed suicide, 1.4 years, compared to 9.5 years when death was due to alcohol. Half of the men who visited the primary health care centre six months before death were single, and the problem of the final antemortem visit anticipated the cause of death in the majority of these men.

Life after divorce: a study of newly divorced middle-aged men in Sweden.

Thirty-two divorced men aged 30 to 44 years answered a questionnaire within seven months after the dissolution of their marriages. They were asked for life circumstances (housing and working conditions, economic situation) and perceived health and a personal interview was also performed with most of the men. Primary health care records during the year of divorce were analysed according to number of visits and reasons for encounter. A majority of the men perceived working conditions as satisfying while about half of the men had moved to less satisfying housing conditions. They recorded a high number of psychological symptoms (insomnia, fatigue, anxiety) but this was not recognizable in the primary health care records. Ten men had had more than one divorce and in this sub-group there were several alcohol abusers. The importance of primary health care in the early recognition of the complex situation of divorced men is stressed.

Modulation of oxotremorine-induced tremor by central beta-adrenoceptors.

The muscarinic agonist oxotremorine was used to induce tremor in rats pretreated with methylatropine. An objective assessment of tremor intensity was accomplished by means of an accelerometer-based recording system. The non-selective, lipophilic beta-adrenoceptor antagonist propranolol dose-dependently suppressed tremor intensity, whereas the R-isomer of propranolol was without effect, verifying beta-adrenoceptor involvement. Since the hydrophilic, non-selective beta-antagonist nadolol was ineffective, the effect appears to be located inside the blood-brain barrier. The beta 2-selective antagonist ICI 118, 551 dose-dependently reduced tremor intensity, whereas selective blockade of beta 1-adrenoceptors with metoprolol had no effect, indicating the participation of a beta 2-adrenoceptor. On the other hand, the lipophilic beta 2-agonist clenbuterol dose-dependently enhanced tremor induced by oxotremorine. Determination of circulating plasma catecholamine concentrations revealed that the effect of beta-antagonists on tremor was not secondary to an effect on the oxotremorine-induced rise in catecholamine levels. Thus, the results suggest that beta 2-adrenoceptors located inside the blood-brain barrier are able to modulate oxotremorine-induced tremor in rats.

Blockade of central beta-adrenoceptors attenuates tremor induced by 5-hydroxytryptamine (5-HT)-receptor activation in rats.

The effect of beta-adrenoceptor antagonists, varying in lipophilicity and receptor selectivity, were studied on tremor elicited by L-5-hydroxytryptophan (L-5-HTP) in rats pretreated with a peripherally acting decarboxylase inhibitor and a monoamine oxidase inhibitor, or by the directly acting 5-HT agonist 5-methoxy-N,N-dimethyltryptamine (5-MeODMT). Plasma levels of the beta-adrenoceptor antagonists were determined simultaneously. The non-selective lipophilic adrenoceptor antagonist propranolol was found to dose-dependently reduce tremor intensity, whereas the non-selective hydrophilic adrenoceptor antagonist sotalol had no effect, indicating a central site of action. Furthermore, beta 1-selective blockade with the adrenoceptor antagonist metoprolol had no effect on tremor intensity, whereas the beta 2-selective antagonist ICI 118,551 dose-dependently suppressed tremor intensity, suggesting that the beta-adrenoceptor subtype involved is of the beta 2-type. These results suggest that blockade of centrally located beta 2-adrenoceptors are able to attenuate the tremor response following 5-hydroxytryptamine receptor activation.

Central beta-adrenoceptors can modulate 5-hydroxytryptamine-induced tremor in rats.

The effects of two beta 2-adrenoceptor agonists with different lipophilicities were studied on tremor induced by L-5-hydroxytryptophan (L-5-HTP) in pargyline- and carbidopa-pretreated rats. Tremor was recorded and analysed by an objective method based on accelerometry. Clenbuterol, a lipophilic beta 2-selective agonist, dose-dependently enhanced tremor intensity, whereas the hydrophilic beta 2-agonist terbutaline had no effect. The clenbuterol-induced enhancement of tremor was completely abolished by the beta 2-selective antagonist ICI 118,551 but unchanged by the beta 1-selective antagonist metoprolol. The results suggest that centrally located beta 2-adrenoceptors can mediate a modulation of 5-hydroxytryptamine-induced tremor in rats.

Beta-adrenoceptor agonists do not cause tremor in the conscious rat.

The ability of some beta-adrenoceptor agonists to induce tremor via skeletal muscle beta 2-adrenoceptors in conscious unrestrained rats has been investigated. Tremor was assessed by visual observation and by an objective method based on accelerometry. Infusion of isoprenaline or terbutaline did not cause tremor, neither did beta-stimulation potentiate an established mild tremor produced by central muscarinic receptor stimulation. Since beta-agonists readily produce tremor in man via skeletal muscle beta 2-adrenoceptor stimulation, our findings indicate that these receptors have a different function in the rat.

Objective quantification of tremor in conscious unrestrained rats, exemplified with 5-hydroxytryptamine-mediated tremor.

To accomplish an objective quantification of tremor in conscious unrestrained rats, a registration device based on accelerometry was developed. Tremor intensity was continuously recorded by a small piezoresistive accelerometer (Egal 125-10D, Entran Devices) mounted on the back of the freely moving rat. The accelerometer was connected to a Grass Polygraph, where the analog signal was quantified as arbitrary marks/min by a 7P10 integrating unit. The integrated signals were then further analyzed by a desk-top computer (Luxor ABC-80). By analyzing tremor response to central serotonergic system activation, the reproductibility as well as the advantages and limitations of this recording system were demonstrated.

Reduced brain serotonergic activity after repeated treatment with beta-adrenoceptor antagonists.

Rats were treated subchronically (14 days) or acutely (single dose) with the beta 1-selective adrenoceptor antagonist metoprolol or the beta 3-selective adrenoceptor antagonist ICI 118,551. Metoprolol (350 mg/kg/day for 14 days, orally) significantly reduced the 5-hydroxytryptophan (5-HTP) accumulation when measured 30 min after inhibition of L-amino acid decarboxylase by NSD 1015 (100 mg/kg IP) in the limbic forebrain, the corpus striatum, the cerebral cortex, the brain stem, and in the cerebellum. ICI 118,551 (0.5 mg/kg, twice daily for 14 days, SC) also significantly reduced the 5-HTP accumulation in the same brain regions except in the corpus striatum and the brain stem Simultaneously assayed tryptophan levels were largely unaffected. Thus sustained beta-adrenoceptor blockade causes a decrease in the in vivo rate of tryptophan hydroxylation in various rat brain regions. The subchronic treatments with metoprolol or ICI 118,551 also significantly reduced the endogenous levels of 5-hydroxytryptamine (5-HT) in the various rat brain regions studied. Acute treatment with either metoprolol (2 mg/kg SC) or ICI 118,551 (0.5 mg/kg SC) did not affect the 5-HTP accumulation or the endogenous 5-HT levels in the brain regions studied. This inhibitory effect on brain 5-HT systems produced by sustained beta-adrenoceptor blockade may be of significance both for the long-tem cardiovascular action and for occasional neuropsychiatric side effects during beta-blocking therapy.