Prävalenz, Ätiologie und Prognose von Husten in der Primärversorgung – eine systematische Übersichtsarbeit symptomevaluierender Studien

Background: Cough is one of the most common presenting conditions in primary care. For evidence-based decision making, primary care physicians need setting-specific knowledge about prevalences, pre-test probabilities of underlying diseases and a symptom’s course. We performed a systematic review and meta-analysis of symptom-evaluating studies on the prevalence/incidence, aetiologies and prognosis of cough in primary care. Data was extracted for adults, patients of all age groups and children. Methods: Following a pre-defined algorithm, MEDLINE and EMBASE were evaluated systematically up to November 2019 resp. January 2020. The reference lists of all relevant papers were screened (snowball search). We included original research articles in English, French, and German, as long as they focused on unselected study populations consulting a primary care physician for cough. Eligibility criteria and methodological quality were assessed independently by two reviewers. We visualized probability estimates and variance between studies with forest plots. If justifiable in terms of heterogeneity, we performed a meta-analysis in subgroups, using the random effects model. Results: 73 references met inclusion criteria. 60 publications (31 studies) reported data on adults or on patients of all age groups, with data on prevalence/incidence of cough in 22 studies, on aetiology in twelve studies and on prognosis in four studies. For children, we identified 20 eligible publications (19 studies): 14 studies on prevalence, five on aetiology and one on prognosis. Prevalence estimates in adults and patients of all age groups were 3.8-4.2% in Western primary care and 10.3.-13.8% in Africa, Asia and South America. In comparison, the corresponding incidence estimates were 12.5% (Western primary care) and 6.3-6.5% (Africa, Asia and South America). Cough in children was more frequent than in adults, with lowest prevalences in adolescents and in summer: Prevalence estimates varied between 4.7 and 23.3% of all reasons for an encounter, or up to estimates of 60% when related to patients or consultations. In Western countries, acute cough in both adults and children was mostly caused by upper respiratory tract infections and bronchitis, followed by influenza and asthma. Potentially serious diseases like pneumonia, COPD, heart failure and suspected malignancy showed low prevalence estimates. For both adults and children, findings on subacute/chronic cough were rare. In children with a cough which persisted more than two weeks, underlying aetiologies were recurrent respiratory tract infection (27.7%), asthma (up to 50.4%) and pertussis (37.2%). In adults and patients of all age groups with acute cough, complete recovery was reported by 40.2- 67% of patients after two weeks, and by 79% after four weeks. Median recovery time was nine to eleven days. About 21.1-35% of patients re-consulted; 0-1.3% of acute cough patients were hospitalized and none died. In children with subacute and chronic cough the total duration of cough ranged from 24 to 192 days. About 62.3% of children suffering from prolonged cough were still coughing two months after the beginning of symptoms. Discussion: Prevalences and incidences of cough are high and show regional variation. Our findings fit in with current guideline recommendations supporting a thoughtful wait-and-see approach in acute cough, as mainly caused by usually self-limiting respiratory tract infections. For chronic cough in children, a special awareness for the possibility of asthma and pertussis is necessary. Further high-quality research on aetiological pre-test probabilities is needed, as setting-specific evidence is rare, especially on the aetiologies of subacute or chronic cough in Western primary care. To improve evidence regarding a wait-and-see strategy vs. the implementation of further diagnostics, referral or treatment, it is important that more prognostic studies, which are based on symptoms, are carried out in primary care. These should contain standardized outcomes, a sufficiently long follow-up period and an unselected study population. Our study presents epidemiological data from before the outbreak of the COVID-19 pandemic. It will be interesting to compare these data to future research results of the post-pandemic era.