Skip to main content Accessibility help
×
Home

Contents:

Information:

  • Access

Actions:

      • Send article to Kindle

        To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Authors' reply
        Available formats
        ×

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Authors' reply
        Available formats
        ×

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Authors' reply
        Available formats
        ×
Export citation

We thank Professor Kawada for his careful reading of our paper. He raises a number of issues that we address here. First, it is important to recognise that, while we used the same study (the English Longitudinal Study of Ageing, ELSA) in each of the papers he describes, 1,2 we asked very different questions of the data.

In the first paper, 2 we were interested in whether depression symptom severity, across the full range of scores measured on a single occasion, was related to later risk of all-cause mortality. If mortality effects are seen in people with mild-to-moderate depression, this could potentially point to the need for treatment at lower levels than is currently the case. Wishing to test whether the dose–response association we found for psychological distress (a combined measure of depression and anxiety) and all-cause mortality in the Health Surveys for England – Scottish Health Surveys collaboration was replicated using a depression-specific inventory, 3 we used an administration of the Center for Epidemiologic Studies Depression Scale (CES-D) during wave 1 of data collection in ELSA. On relating those scores (higher score indicated greater depression severity) to mortality experience over the following 9 years, after basic statistical adjustments, there was a stepwise effect that, as Kawada indicates, seemed to plateau in people reporting the most severe symptoms.

In the second paper, recognising that symptoms of depression tend to be, as Kawada points out, time-varying, 4,5 in order to better capture depression exposure we capitalised on the serial measurements made over 4 waves of data collection (8 years) in ELSA, which is rare in population-based studies. We found that the number of waves a study member was denoted as a ‘case’ (based on a score of ⩾3 on the CES-D) was positively associated with deaths occurring after the final wave of data capture. Again, we found evidence of a dose–response effect. Importantly, in both papers, taking into account differences across the depression groups in levels of physical activity, cognitive function, functional impairments and physical illness led to complete attenuation of any relationships. This suggests that these factors either confound or mediate the depression–mortality gradient.

Contrary to Kawada's view, we do not think that studies of sick populations – in the examples given, patients with cardiovascular disease (CVD) – offer any insight into the link between depression and the future occurrence of CVD events (aetiology). In our papers, concerns regarding a lack of statistical power (for analyses of depression duration) and space constraints (for analyses of depression severity) prevented us from reporting relationships with cause-specific mortality, including CVD. We take this opportunity to do so here. Figure 1 shows the analysis requested by Kawada. We see associations of the duration of depression symptoms (Fig. 1a: 233 CVD deaths in 9560 people over a median of 3.6 years of follow-up adjusted for age and gender) and the severity of symptoms (Fig. 1b: 703 CVD deaths in 11 104 people over a median of 9.7 years of follow-up adjusted for age, gender and ethnicity) with CVD mortality. These figures show a somewhat similar shape to that apparent for all-cause mortality for the association with duration of depressive symptoms and a flatter association for symptom severity. In both analyses the wide confidence intervals illustrate the low precision of the point estimates.

Fig. 1 Hazard ratios (95% confidence intervals) for the associations of duration (a) and severity of depressive symptoms (b) with cardiovascular disease mortality in the English Longitudinal Study of Ageing. HRs adjusted for age and gender (a); age, gender and ethnicity (b).

In conclusion, our results seem to accord with extant literature that has found basic adjustments reveal effects are lost after taking into account multiple covariates. Advancing this field now requires a more rigorous examination of cause and effect. Among other approaches, this could be tested using aetiological trials in which the impact of successful treatment for depression on CVD risk is quantified, or using Mendelian randomisation where gene variants for depression are employed as instrumental variables to explore apparently unconfounded associations with CVD.

1 White, J, Zaninotto, P, Walters, K, Kivimäki, M, Demakakos, P, Biddulph, J, et al. Duration of depressive symptoms and mortality risk: the English Longitudinal Study of Ageing (ELSA). Br J Psychiatry 2016; 208: 337–42.
2 White, J, Zaninotto, P, Walters, K, Kivimäki, M, Demakakos, P, Shankar, A, et al. Severity of depressive symptoms as a predictor of mortality: the English longitudinal study of ageing. Psychol Med 2015; 45: 2771–9.
3 Russ, TC, Stamatakis, E, Hamer, M, Starr, JM, Kivimäki, M, Batty, GD. Association between psychological distress and mortality: individual participant pooled analysis of 10 prospective cohort studies. BMJ 2012; 345: e4933.
4 Judd, LL, Akiskal, HS. Delineating the longitudinal structure of depressive illness: beyond clinical subtypes and duration thresholds. Pharmacopsychiatry 2000; 33: 37.
5 Judd, LL, Akiskal, HS, Maser, JD, Zeller, PJ, Endicott, J, Coryell, W, et al. A prospective 12-year study of subsyndromal and syndromal depressive symptoms in unipolar major depressive disorders. Arch Gen Psychiatry 1998; 55: 694700.